UNLOCKING
AMERICA

Why and How to Reduce
America’s Prison Population
NOVEMBER 2007

PUBLISHED BY:
THE JFA INSTITUTE | 5 Walter Houp Ct, NE | Washington, DC 20002 | www.jfa-associates.com

UNLOCKING
AMERICA

Why and How to Reduce
America’s Prison Population

NOVEMBER 2007
WWW.JFA-ASSOCIATES.COM

Authors
James Austin, President, The JFA Institute
Todd Clear, Professor, John Jay College of Criminal Justice
Troy Duster, Professor, New York University
David F. Greenberg, Professor, New York University
John Irwin, Professor Emeritus, San Francisco State University
Candace McCoy, Professor, City University of New York
Alan Mobley, Assistant Professor, San Diego State University
Barbara Owen, Professor, California State University, Fresno
Joshua Page, Assistant Professor, University of Minnesota

Acknowledgments
Many people made important contributions to this report either by providing data, reviews of earlier
reports, and editing of the final report. They include Patricia Hardyman, Charles Jones, Frank Zimring,
Ron Huff, Michael Jacobson, Marvin Mutch, Keith Hardison, Judith Green, Eric Sterling, Gene
Guerrero, Nkechi Taifa, Susan Tucker, Aurie Hall, Robert Schwartz, Debra Rubino, Belinda Cooper,
Amy Solomon, Rachel Barkow, Jodi Lewen, and Frank Cullen.

Funding provided by:
THE ROSENBAUM FOUNDATION

and
OPEN SOCIETY INSTITUTE

Designed by:
Darcy Harris
www.herdesignereye.com

iv

UNLOCKING AMERICA

Why and How to Reduce America’s Prison Population

TABLE OF CONTENTS

1
Prologue
3
I. Crime Rates and Incarceration
11
II. Three Key Myths about Crime and Incarceration
15
III. The Limits of Prison-Based Rehabilitation and Treatment Programs in Reducing the Prison Population
18
IV. Decarceration, Cost Savings, and Public Safety
20
V. Six Recommendations
29
VI. Concluding Remarks

v

Unlocking America

Prologue
“Mr. Libby was sentenced to 30 months of prison, two years of probation and a $250,000
fine…I respect the jury’s verdict. But I have concluded that the prison sentence given to Mr. Libby
is excessive.”—President George W. Bush. July 2, 2007.
President Bush was right. A prison sentence for Lewis
“Scooter” Libby was excessive—so too was the long three
year probation term. But while he was at it, President
Bush should have commuted the sentences of hundreds
of thousands of Americans who each year have also received prison sentences for crimes that pose little if any
danger or harm to our society.
In the United States, every year since 1970, when only
196,429 persons were in state and federal prisons, the
prison population has grown. Today there are over 1.5 million in state and federal prisons. Another 750,000 are in
the nation’s jails. The growth has been constant—in years
of rising crime and falling crime, in good economic times
and bad, during wartime and while we were at peace. A
generation of growth has produced prison populations
that are now eight times what they were in 1970.
And there is no end to the growth under current policies. The PEW Charitable Trust reports that under current
sentencing policies the state and federal prison populations will grow by another 192,000 prisoners over the
next five years. The incarceration rate will increase from
491 to 562 per 100,000 population. And the nation will
have to spend an additional $27.5 billion in operational
and construction costs over this five-year period on top of
the over $60 billion now being spent on corrections each
year.1
This generation-long growth of imprisonment has
occurred not because of growing crime rates, but because
of changes in sentencing policy that resulted in dramatic
increases in the proportion of felony convictions resulting
in prison sentences and in the length-of-stay in prison that
those sentences required. Prison populations have been
growing steadily for a generation, although the crime rate
is today about what it was in 1973 when the prison boom
started. It is tempting to say that crime rates fell over the
past dozen years because imprisonment worked to lower
them, but a look at data about crime and imprisonment
will show that prison populations continued to swell long
after crime rates declined and stayed low. Today, whatever

1

is driving imprisonment policies, it is not primarily crime.
Prisons are self-fueling systems. About two-thirds
of the 650,000 prison admissions are persons who have
failed probation or parole—approximately half of these
people have been sent to prison for technical violations.
Having served their sentences, roughly 650,000 people
are released each year having served an average of 2-3
years. About 40% will ultimately be sent back to prison
as “recidivists”—in many states, for petty drug and property crimes or violations of parole requirements that do
not even constitute crimes. This high rate of recidivism is,
in part, a result of a range of policies that increase surveillance over people released from prison, impose obstacles
to their reentry into society, and eliminate support systems that ease their transition from prison to the streets.
Prison policy has exacerbated the festering national
problem of social and racial inequality. Incarceration rates
for blacks and Latinos are now more than six times higher
than for whites; 60% of America’s prison population is either African-American or Latino. A shocking eight percent
of black men of working age are now behind bars, and
21% of those between the ages of 25 and 44 have served
a sentence at some point in their lives. At current rates,
one-third of all black males, one-sixth of Latino males, and
one in 17 white males will go to prison during their lives.
Incarceration rates this high are a national tragedy.2
Women now represent the fastest growing group of
incarcerated persons. In 2001, they were more than three
times as likely to end up in prison as in 1974, largely due
to their low-level involvement in drug-related activity and
the deeply punitive sentencing policies aimed at drugs.
The massive incarceration of young males from mostly
poor- and working-class neighborhoods—and the taking of women from their families and jobs—has crippled
their potential for forming healthy families and achieving
economic gains.
The authors of this report have spent their careers
studying crime and punishment. We are convinced that
we need a different strategy. Our contemporary laws

and justice system practices exacerbate the crime problem, unnecessarily damage the lives of millions of people,
waste tens of billions of dollars each year, and create less
than ideal social and economic conditions in many sections of our largest American cities.
This report focuses on how we can reduce the
nation’s prison population without adversely affecting
public safety. For this to happen, we will need to reduce
the number of people sent to prison and, for those who
do go to prison, shorten the length of time they spend
behind bars and under parole and probation surveillance. People who break the law must be held accountable, but many of those currently incarcerated should
receive alternative forms of punishment, and those who
are sent to prison must spend a shorter period incarcerated before coming home to our communities. Our recommendations would reestablish practices that were
the norm in America for most of the previous century,
when incarceration rates were a fraction of what they
are today.
We first summarize the current problem, explaining how some of the most popular assumptions about
crime and punishment are incorrect. In particular, we

demonstrate that incarcerating large numbers of people
has little impact on crime, and show how the improper
use of probation and parole increases incarceration
rates while doing little to control crime. We then turn
to ideas about how to change this flawed system. We
set out an organizing principle for analyzing sentencing
reform, embracing a retributive sentencing philosophy
that is mainstream among contemporary prison policy
analysts and sentencing scholars.
Based on that analysis, we make a series of recommendations for changing current sentencing laws and
correctional policies. Each recommendation is practical and cost-effective. As we show through examples of
cases in which they have been tried, they can be adopted without jeopardizing public safety. If implemented
on a national basis, our recommendations would gradually and safely reduce the nation’s prison and jail populations to half their current size. This reduction would
generate savings of an estimated $20 billion a year that
could then be reinvested in far more promising crime
prevention strategies. The result would be a system of
justice and punishment that is far less costly, more effective, and more humane than what we have today.

“Punishment Does Not Fit the Crime” Some Recent Examples
“Offenders”

Prior Record

Crime

Description

Prison Sentence

Elisa Kelly
George Robinson
Mother and stepfather3

None

Nine counts of contributing
to the delinquency of a
minor

Hosting drinking party for son’s
nine friends at parent’s home

Original sentence
of 8 years- later
reduced to 27
months

Cecilia Ruiz
Single parent – two children
ages 6 and 84

None

Forgery

Deleting a DUI conviction from
the county DUI data base

42 months

Jessica Hall
Unemployed mother of three
children with Marine husband
serving in Iraq5

None

Throwing a missile at an
occupied vehicle

Threw a cup of McDonald’s
coffee at another car that cut her
off while driving

24 months

Lewis “Scooter” Libby6

None

Perjury

Provided false testimony to U.S.
Attorney (four counts)

30 months

Stephen May7

None

Child Molestation

Inappropriately touched two girls
and a boy – there was no sexual
activity or penetration

75 years

Genarlow Wilson8

None

Aggravated child
molestation

17 year old male had consensual
oral sex with a 15 year old girl at
a party that was video taped.

10 years

Public Safety, Public Spending: Forecasting America’s Prison Population 2007-2011. Philadelphia, PA: Pew Charitable Trusts, 2007.
In some cites the percentages are higher. For example, in Baltimore, one in five young black men between ages 20 and 30 is incarcerated, and 52% are under some form
of correctional supervision. Jason Ziedenberg and Eric Lotke. Tipping Point: Maryland’s Overuse of Incarceration and the Impact on Public Safety. Washington, DC: Justice
Policy Institute, 2005.
3
Washington Post, July 4, 2007, pages A1, A11, “Penalties for Teen Drinking Parties Vary Widely in Area”.
4
Washington Post, June 30, 2007, page B4, Ex-Aide Given 3 ½-Year Sentence.
5
www.washingtonpost.com/wp-dyn/content/article/2007/02/17/AR2007021701560.htt
6
Washington Post, July 4, 2007, page A4, “Bush Says He’s Not Ruling Out Pardon for Libby”.
7
State of Arizona vs. Stephen May, Maricopa Superior Court, No. CR2006-030290-001 SE.
8
abcnews.go.com/Primetime/LegalCenter/story?id=1693362&page=1.
1
2

2

I

Crime and Incarceration

By far the major
reason for the
increase in the prison
populations at least
since 1990 has been
longer lengths of
imprisonment.

WHY THE PRISON EXPLOSION?
America’s incarceration rate is exploding. In 1970, there were fewer than 200,000
people in prison. By 2006, there were approximately 1.6 million state and federal
prisoners (or nearly 500 per 100,000 population). Each year over 730,000 people are
admitted to state and federal prisons, and
a much larger number (over 10 million) go
to local jails. If we add to the prison population the nearly 750,000 people incarcerated in local jails today—at the beginning
of 2007—the total number imprisoned in
the U.S. on any given day is 2.2 million.1
The United States is the world leader in imprisonment. China, with a much
larger population, has the second largest
incarcerated population, with 1.5 million
imprisoned. With 737 people incarcerated per 100,000 persons, the U.S. also leads
the world in rates of incarceration—well
above Russia, which has the next highest rate of 581 per 100,000.2 The other

Table 1: Sentence Lengths, Time Served, and Length of Parole
Supervision, 1993 versus 2002

Length of Supervision

1993

2002

Average Sentence

66 months

65 months

Average Time Served

21 months

30 months

Average Parole Supervision

19 months

26 months

Total Time Under Supervision

40 months

56 months

Source: Prison Statistics. US Department of Justice, Office of Justice Programs, Bureau of Justice
Statistics.1 Aug. 2006. http://www.ojp.usdoj.gov/bjs/prisons.htm

Western democratic countries manage
with prison populations far smaller than
ours.
By far the major reason for the increase
in prison populations at least since 1990
has been longer lengths of imprisonment.
The adoption of truth in sentencing provisions that require prisoners to serve most
of their sentences in prison, a wide variety
of mandatory minimum sentencing provisions that prevent judges from placing
defendants on probation even when their
involvement in the conduct that led to
the conviction was minor, reductions in
the amount of good time a prisoner can
receive while imprisoned, and more conservative parole boards have significantly
impacted the length of stay.
For example, in a special study by the
U.S. Department of Justice on truth in
sentencing, between 1990 and 1997, the
numbers of prison admissions increased
by only 17% (from 460,739 to 540,748),
while the prison population increased
by 60% (from 689,577 to 1,100,850). This
larger increase in the prison population
can only be caused by a longer length of
stay.3
This is further confirmed in Table
1, which shows sentence lengths, time
served, and period spent on parole supervision. These U.S. Department of Justice
data are based on individuals released in
1993 and 2002. The 2002 data underestimate the average length of current prison
sentences because they do not include
time served by prisoners sentenced under

1
William J. Sabol, Todd D.Minton, and Paige M.Harrsion. Prison and Jail Inmates at Midyear 2006. Bureau of Justice Statistics
Bulletin, Washington, DC: US DOJ,Office of Justice Programs, 2007.
2
Prison Brief—Highest to Lowest Rates. International Centre for Prison Studies, King’s College, London. 2006 http://www.kcl.
ac.uk/depsta/rel/icps/world/
3
Paula M. Ditton and Doris James Wilson. “Truth in Sentencing in State Prisons.” Washington, DC: US DOJ, Bureau of Justice
Statistics, 1999.

3

recent punitive laws (such
Table 2: Economic Loss to Victims and Costs of Incarceration
as “three strikes and you’re
out”) who have not yet been
Prison
Average
Prison
Pretrial
Total
released. Nevertheless, the
Time
Incarceration
Type of Crime
Victim
Sentence
Time
Time
Served
Costs*
average time served by those
Loss
(in mos)
(in mos)
(in mos)
(in
mos)
who were released still in$1,258
94
5
55
60
$113,000
creased substantially—from Robbery
21 to 30 months. Similarly, Burglary
$1,545
52
5
29
34
$64,000
the parole supervision pe$730
34
5
20
25
$47,000
riod increased from 19 to Larceny Theft
$6,646
27
5
17
22
$41,000
24 months, and the total Auto Theft
average period of supervi- Source: Matthew R. Durose and Patrick A. Langan. Felony Sentences in State Courts, 2000. Washington, DC: Bureau of
sion increased from 40 to 56 Justice Statistics, 2003, and Callie M. Rennison and Michael Rand. Criminal Victimization, 2002. Washington, DC: Bureau of
Justice Statistics, 2003.
months.
Not only are our lengths
spend over $200 billion each year to fund the criminal
of imprisonment significantly
justice system.6 In contrast, the total economic loss to viclonger than they were in earlier periods in our penal histotims of crime in 2002 was an estimated $15.6 billion, or
ry, but they are considerably longer than in most Western
about one-tenth of the total cost of the nation’s criminal
nations. For the same crimes, American prisoners receive
justice system.7 The typical (median) costs per crime for
sentences twice as long as English prisoners, three times
each victim was $100, which includes losses from propas long as Canadian prisoners, four times as long as Dutch
erty theft or damage, cash losses, medical expenses, and
prisoners, five to 10 times as long as French prisoners, and
lost pay. While the financial losses and physical and emo4
five times as long as Swedish prisoners. Yet these countional injuries sustained by victims can be significant, they
tries’ rates of violent crime are lower than ours, and their
represent only a fraction of what it costs to incarcerate the
rates of property crime are comparable.5
offenders.
Most prisoners are incarcerated for crimes that do
Table 2 illustrates the vast disparity between the
not compare with the costs of their imprisonment. We
economic losses associated with four common crimes
and the amount expended
to incarcerate the offender.
Figure 1: Crime and Incarceration Rate Comparisons
For example, the average
loss associated with a robbery reported to the police is $1,258. The typical
prison sentence for robbery in the United States is
94 months, or about eight
years, of which the typical
time served is 55 months.
Together with the time
spent in jail pre-trial, the average robbery offender is
incarcerated for 60 months
at a cost of approximately
$113,000.
This historic rise in incarceration has often been
attributed to the “fact” that
David P. Farrington, Patrick A. Langan, and Michael Tonry, eds. Cross-National Studies in Crime and Justice. Washington, DC: US DOJ, Bureau of Justice Statistics, 2004.
Franklin E. Zimring and Gordon Hawkins, Crime Is Not the Problem: Lethal Violence in America. New York: Oxford UP, 1997.
6
http://www.ojp.usdoj.gov/bjs/eande.htm
7
Callie M. Rennison and Michael Rand. Criminal Victimization, 2002. Washington, DC: Bureau of Justice Statistics, 2003.
4
5

4

Unlocking America

in the early 1970s, the U.S. faced a steadily increasing
crime problem, leaving no choice but to increase the use
of incarceration massively. But this explanation for the imprisonment binge is misleading and incomplete. Crime
rates have grown in other countries and states within the
U.S. without provoking a large growth in prison populations. There are various ways a country can respond to increased crime; more prisons is just one of them. Moreover,
statistics show that it was not primarily a rise in crime that
fueled the increase in incarceration rates.
Figure 1 compares changes in the nation’s crime
rates—as measured by the FBI’s Uniform Crime Reports
(UCR) that are based on police records that primarily reflect victim reports to their local police departments—
with changes in the rates of incarceration from 1931 to
2004. After being relatively stable for decades, between
1964 and 1974 the UCR Crime Index—which measures
murder, assault, rape, burglary, robbery, theft, auto theft,
and arson—increased significantly even as the incarceration rate remained relatively stable.8 Thereafter, the UCR

crime rate swung up and down until 1992 when it began
a steady decline.
The UCR is one of two ways that the government tracks
crime rates. The other, the National Crime Victimization
Survey (NCVS), was established in 1973 in recognition
of the fact that the UCR had major biases in the way it
gauged crime. The NCVS is often considered a more accurate and complete picture of crime in the U.S., as it is
based on interviews of household members.
As shown in Figure 2, NCVS shows no increases in
property crimes from 1973 to 1980. Similarly, NCVS show
no increase in violent crimes between 1973 and in 1993,
unlike the UCR data which showed a steady increase over
the same time period. There is contradictory evidence
of a 1970s crime epidemic, which was a major rationale
for expanding the use of imprisonment.9 Regardless of
which measuring method is used, there was no large drop
in the incarceration rate before crime rates began to increase after 1963. While a variety of theories exist as to
why crime rates rose from 1964 to 1974, the cause was not
a large drop in prison populations, which remained
fairly stable until 1973.10
Figure 2: UCR and NCVS Crime Rate Comparisons
Prison populations began
increasing only after crime
rates had already stabilized
or, according to the crime
victimization surveys, had
already began to decline.
Beginning in the
1960s, law and order advocates declared a war on
crime. Conservative politicians (starting with Barry
Goldwater in the early
1960s), backed by religious
groups (the Moral Majority
of the 1970s) and rightwing media figures (such
as talk-radio hosts of the
1980s), argued that crime
was out of control largely
because lenient judges
gave lawbreakers too

8
Prison populations fell by about eight percent between 1960 and 1970. This small drop could not remotely have been a major cause of the 2.4-fold rise in UCR index
offenses in that decade. Sourcebook of Criminal Justice Statistics. US DOJ, Bureau of Justice Statistics. 278, 500 http://www.albany.edu/sourcebook/
9
Criminologists debate the sources of the discrepancy between the growth in violent crime shown in UCR data and the decline shown by the NCVS. The U.S. Department
of Justice has suggested that victims were more likely to report crimes to the police, and law enforcement agencies improved their recording and reporting systems. If these
explanations are correct, these changes in reporting created the illusion of a growing crime problem when there actually was none. Also note that the NCVS property rate is
based on the number of households, which is lower than the number of persons.
10
As the NCVS was started in 1973, it cannot be used to determine if—or how much—crime rates increased between 1964 and 1973. Most criminologists agree, however,
that there was an increase, and that some of the increase was due to demographics–the large numbers of baby boomers passing through their high crime-committing years.
People aged 15 to 24 commit a substantial proportion of index crimes. The persons in the baby-boom cohorts started reaching 15 in 1964, and began turning 40 around 1990.
But demographics cannot explain all of the increase.

5

Why and How to Reduce America’s Prison Population

many chances before they
were punished, predatory
Figure 3: Incarceration Rates by Crime Category (1980-2003)
criminals avoided punishment because of technicalities in the law, and criminals
returned to the streets after
serving short sentences.
These advocates of a war
on crime suggested harsh
mandatory punishments
were needed that would
both incapacitate people
convicted of serious crimes
and deter others from
breaking the law.
The media contributed
to the fervor over the “crime
problem” through its unrelenting focus on crime—the
more heinous and sensational, the better. Broadcast
journalists discovered that
sensational crimes drew
Remedying social inequities, they maintained, was
large audiences. The media
not the job of the government and would not reduce
publicized horrific but rare crimes (the kidnapping and
crime. Their solution—more people in prison—rested on
murder of Polly Klass, the sexual murder of Megan Kanka,
the use of an institution that had become substantially
and the attack perpetrated by Willie Horton while on work
discredited among penologists in the 1960s. Their views
release). In so doing, it led Americans to believe wrongly
were enthusiastically supported by the federal governthat they were at high risk of being assaulted, raped, or
ment, which subsequently funded studies to identify the
murdered.
“career criminal.”
During this same period, a growing number of neoPublic receptivity to this hawkish perspective was
conservative social scientists claimed that “law and order”
enhanced by anxieties associated with the rapid rise in
ideas and policies rested on a sound scientific foundation.
crime during the 1960s, the eruption of riots in a number
In 1975, political scientist James Q. Wilson published the
of American cities, and the advent of the civil rights and
widely read Thinking About Crime, in which he soberly
feminist movements, challenging longstanding racial and
admonished liberal/humanitarian social scientists to regender hierarchies. These developments, exploited by
member that “...wicked people exist. Nothing avails except
well-funded conservative polemicists whose arguments
to set them apart from innocent people.”11
were widely publicized in the media, helped to shape
For Wilson and those who initially shared his views,
public response to the high-profile death of athlete Len
incapacitating what they believed was a growing and
Bias from a cocaine overdose in 1986. Along with the
more menacing type of criminal was a necessary weapon
growing use of crack cocaine, his death sparked a rapid
in the war against crime. By insisting that only steppedgovernment response. All these factors converged, creatup criminal justice interventions could cope with crime,
ing a “perfect storm” that drove the imprisonment binge.
Wilson and others like John J. DiIulio and William Bennett
Thus after 1975 many laws were passed, supported by
were arguing against the strategies recommended by the
politicians of both parties, designed to increase the probriot commissions of the 1960s and by experts on povability of a prison sentence rather than jail or probation,
erty and racial problems: alleviating poverty, combating
dramatically increasing the length of prison sentences for
discrimination, and opening up opportunities that had
certain crimes, and requiring prisoners to serve a greater
previously been shut off by racial discrimination and class
proportion of their sentences—three results that we will
inequality.
11

James Q. Wilson. Thinking About Crime. New York: Random House, 1975. 235. See also by Wilson: “Lock ‘Em Up.” New York Times Magazine 9 March 1975: 11.

6

Unlocking America

discuss in greater detail below.
Priests are charged with child molestation. Doctors bill
New legislation increasing the penalties for the sale
Medicare for procedures they did not perform. Each day
and possession of cocaine, particularly crack cocaine, inthousands drive while legally drunk, with sometimes-fatal
tensified the upward trend. As is evident in Figure 3, inconsequences.12 Self-report surveys find that lawbreaking
creased penalties for drug offenses strongly fueled much
is widely distributed among the young, especially males.
of the growth of state prison populations starting in the
As many as one-third of all boys are arrested before reachlate 1980s. Since then most of the growth in prison sysing adulthood. A majority of males will be arrested for a
tems has been due to the increased incarceration of pernon-traffic offense at some point in their lives.13
sons for violent crimes (mostly robbery and assaults) and
But given that most of us commit some type of crimes
behavior related to “public order.” Once begun, prison exin our lifetimes, the most severe punishments are targeted
pansion has continued under its own momentum, oblivitoward lower class citizens. It is this class of people we are
ous to any obvious need and unrelated to any useful sowilling to punish disproportionately to their criminal acts.
cial function, at large social
cost. Though penologists
Table 3: If Black and Latino Prisoners
have called attention to the
Had the Same Jail and Prison Incarceration Rates as White Prisoners
failures of mass imprisonMale
Female
ment, the expansion has
Current
Prison Population If Same
Race
Incarceration
Incarceration
taken on a life of its own.
Prisoners
Incarceration Rates as Whites
Rates
Rates
Along
the
way,
Americans have become
Whites
731,200
681
75
731,200
accustomed to demonizing
certain people who break
Blacks
899,200
4,834
352
125,773
the law. This is somewhat
odd, given the fact that
Latino
392,200
1,778
148
136,540
breaking the law is neither
a rare nor demonic act.
Total
2,022,600
993,513
Few citizens go through
life without ever violating a
law.
This willingness to punish them is rooted in our culture
A significant number of Americans cheat on their taxof individualism, which holds that we are free-willed and
es, steal from their employers, receive stolen goods, purfully responsible for our acts. This belief speaks to our tenchase illegal cable boxes, illegally download music, use ildency to assign personal blame for every disapproved act.
legal drugs, or participate in many other illegal acts. Hardly
The truth is that many factors, particularly race and ecoa day goes by without a newspaper story about corporate
nomic status, affect our life situations and limit or expand
executives indicted for fraud, insider trading, or price fixour available choices.
ing. Politicians are arrested and convicted for soliciting
Consider young people with college-educated paror accepting bribes. The most prestigious Wall Street firms
ents. Middle and upper-class parents impart values of
have engaged in large-scale illegal trading practices. Some
achievement through education to their children, arrange
of our most successful and idolized entertainment figures
for their admission into better schools, help them get into
are indicted for shoplifting, domestic violence, drunk drivgood colleges, and support them after graduation while
ing, possessing drugs, and molesting children. Policemen
they attempt to start their “careers.” Their children have
are filmed using excessive use of force on citizens, make
a much higher chance of succeeding in America’s econoarrests based on racial stereotypes, deal in drugs, plant
my than youth from poor, welfare dependent, and often
evidence on innocent people, and lie under oath. Sports
broken families who are exposed to a much different life,
figures are accused of rape, assault and taking steroids.
including dangerous public housing buildings and dys12
In 2002, an estimated 1.5 million drivers were arrested for driving under the influence of alcohol or narcotics—surely more escaped arrest. In 2004, the National Center for
Statistics and Analysis reported that just over 17,000 people died in alcohol-related crashes in 2003. This is slightly more than the 16,503 murdered or who died from nonnegligent homicide in the same year.
13
In a study of a Philadelphia male birth cohort, 35 percent of the youths had at least one incident of contact with the police before age 18. These 3,475 boys with a contact
were linked to 10,214 offenses, including 2,786 index crimes (Marvin E. Wolfgang, Robert M. Figlio, and Thorsten Sellin. Delinquency in a Birth Cohort. Chicago: University of
Chicago Press, 1972. 65, 68-69. By age 30, 47% had at least one police contact for a non-traffic offense (James J. Collins.Offender Careers and Restraint: Probabilities and
Policy Implications. Washington, DC: US DOJ, Law Enforcement Assistance Administration,1977). In a similar study carried out in Racine, Wisconsin, 59-69% had a non-traffic
police confrontation by the same age. Joan Petersilia. “Criminal Career Research: A Review of Recent Evidence.” Crime and Justice: An Annual Review of Research 2 (1980):
321-79.

7

Why and How to Reduce America’s Prison Population

functional public schools. For many of these youth, the
expectation or norm is to drop out of high school and end
up hanging out in neighborhoods filled with other undereducated, unemployed young people.14 Rather than going to college they are headed toward prison.
Where lawbreakers differ culturally and socially, others are less able to empathize with their life circumstances. Middle- and upper-class people understand that they
themselves, their families, and close associates are not defined entirely by their law violations. But they often lack
understanding for those whose life circumstances are different from their own.
The demonization of criminals has become a special
burden for young black males, of whom nearly one-third
will spend time in prison during their life.15 The fear of black
men and other factors fuel the racially disproportionate
imprisonment and convinces many Americans that black
males are an especially “dangerous” class of people, different from the rest of us, the so-called “law-abiding.”16
Today approximately 60% of those incarcerated today are
black or Hispanic.17 In effect, the imprisonment binge created our own American apartheid.
To illustrate the significance of race in our incarceration policies, we estimated the size of today’s prison and
jail population assuming blacks and Latinos had the same
incarceration rate as their white counterparts. The results
are striking. As shown in Table 3, if blacks and Latinos had
the same rates of incarceration as whites, today’s prison and
jail populations would decline by approximately 50%. While
we may differ on the basis for racial differences in imprisonment, there can be no disagreement that this is not the
sign of a healthy society.

DID PRISON EXPANSION CUT CRIME?
Proponents of prison expansion have heralded this
growth as a smashing success.18 But a large number of
studies contradict that claim. Most scientific evidence
suggests that there is little if any relationship between
fluctuations in crime rates and incarceration rates. In
many cases, crime rates have risen or declined independent of imprisonment rates. New York City, for example,
has produced one of the nation’s largest declines in
crime in the nation while significantly reducing its jail and
prison populations.19 Connecticut, New Jersey, Ohio, and
Massachusetts have also reduced their prison populations
during the same time that crime rates were declining.20
A study of crime and incarceration rates from 1980 to
1991 in all 50 states and the District of Columbia shows
that incarceration rates exploded during this period. The
states that increased incarceration rates the least were just
as likely to experience decreases in crime as those that
increased them the most.21 The study’s authors pointed
out that although San Francisco and Alameda counties in
California reduced the number of individuals sentenced
to state prison, their crime rates dropped as much if not
more than those California counties that increased their
prison commitments.
Other studies reach similar conclusions, finding “no
consistent relationship between incarceration rates and
crime rates”22 and “no support for the ‘more prisoners, less
crime’ thesis.”23 One study discovered an initial decrease
in crime related to increases in rates of incarceration, but
no decrease from further increases in incarceration.24
Focusing on California, whose incarcerated population more than tripled during the 1980s, Franklin Zimring

14
See Pierre Bourdieu and Jean Claude Passeron. Reproduction in Education, Society, and Culture. London: Sage Publications, 1990, and, Jay MacLeod. Ain’t No Makin’ It:
Aspirations & Attainment in a Low-Income Neighborhood. Boulder, CO: Westview. 1995. These books address the issues of transmission of cultural capital and success.
15
Prevalence of Imprisonment in the U.S. Population, 1974-2001. Washington, DC: US DOJ, Bureau of Justice Statistics, 2003.
16
Controlling for crime rates, prison populations have grown faster in states with more black residents. Katherine Beckett and Bruce Western. “Governing Social Marginality:
Welfare, Incarceration, and the Transformation of State Policy.” Punishment and Society 3 (2001): 43-59; David F. Greenberg and Valerie West. “State Prison Populations and
Their Growth, 1971-1991,” and David F. Greenberg, “’Justice’ and Criminal Justice.” Criminology 39.3 (2001): 615-53. Darnell F. Hawkins, Samuel L. Myers Jr., and Randolph
N. Stone, eds. Crime Control and Social Justice: The Delicate Balance. Westport, CT: Greenwood Press, 2003.
This is not to deny that for UCR crimes, rates of black involvement are higher than rates of white involvement. However, these disparities do not fully explain the racial
disparities in prison populations, or their growth.
17
Prisoners in 2004. US DOJ. Washington, DC: Bureau of Justice Statistics, 2005.
18
John DiIulio. “Prisons Are a Bargain, by Any Measure.” New York Times 26 Jan. 1996: A19.
19
Between 1993 and 2001, violent crime in New York decreased by 64%, and homicides by 69% while its jail population dropped by 25%, and the number of people
sentenced to prison fell by 42%. Some analysts attribute the dramatic reductions in serious crime and felony arrests to the introduction of new methods of policing, particularly
Compstat. However, San Diego, which also reduced its commitments to prison, experienced comparable declines in crime without using New York’s methods (Michael
Jacobson. Downsizing Prisons: How to Reduce Crime and End Mass Incarceration. New York: NYUP, 2005. 37, 113, 125-27.
20
Ibid., 128.
21
Irwin and Austin, loc. cit., 154-56. For an updated version of this type of analysis, see: Jenni Gainsborough and Marc Mauer. Diminishing Returns: Crime and Decarceration
in the 1990s. Washington, DC: The Sentencing Project, 2000.
22
Michael Lynch. “Beating a Dead Horse: Is There Any Basic Empirical Evidence of the Deterrent Effect of Imprisonment.” Crime, Law and Social Change 31.4 (1999): 361.
Lynch examined data on U.S. crime and imprisonment trends from 1972 through 1993.
23
Tomislav V. Kovandzic and Lynne M. Vieratis. “The Effect of Country-Level Prison Population Growth on Crime Rates.” Criminology & Public Policy 5.2 (2006): 234. The
authors examined the effect of incarceration on crime rates in different Florida counties.
24
Raymond Liedka, Anne Morrison Piehl, and Bert Useem. “The Crime-Control Effect of Incarceration: Does Scale Matter?” Criminology & Public Policy 5.2 (2006): 245-76.
This study analyzed the data on crime rates and incarceration rates for all 50 states and the District of Columbia in the period from 1972 to 2000. The authors believe that
whatever gains were achieved in the dramatic rise in imprisonment that began in the 1970s, there was a diminishing effect over time.

8

Unlocking America

and Gordon Hawkins concluded that this remarkable expansion was paralleled by a reduction in crime of about
15%. Almost all this decrease was in burglary and larceny.
For the other offenses the reduction was “weak to negligible.” The role of the prison expansion in bringing about
even that 15% was doubtful, because arrest statistics
showed that the drop occurred mainly for juveniles, who
were less likely than adults to be locked up.25 A contrary
view has been offered by some analysts, such as William
Spelman, who argues that the crime rate today would be
25% higher were it not for the large increases in imprisonment from 1970 to 1990.26 Assuming for the sake of
discussion that this figure is correct, it is not a large effect, considering the enormous increase in imprisonment
needed to achieve it. However, Spelman’s estimate is probably too large. His analysis is based on national trends and
does not explain why some states and counties that lowered their incarceration rates experienced the same crime
reductions as states that increased incarceration; and his
estimates of crime reductions rely on controversial data in
the form of prisoners’ own claims about how much crime
they had committed before incarceration. More significantly, Spelman agrees that further investment in expanding the prison population will have little if any impact on
crime rates.
More recent estimates based on individual states and
counties within states have estimated the crime-reduction impact of prison growth to be much smaller or nonexistent.27 Research on crime and incarceration does not
consistently indicate that the massive use of incarceration
has reduced crime rates.
In sum, studies on the impact of incarceration on
crime rates come to a range of conclusions that vary from
“making crime worse” to “reducing crime a great deal.”
Though conclusive evidence is lacking, the bulk of the evidence points to three conclusions: 1) The effect of imprisonment on crime rates, if there is one, is small; 2) If there is
an effect, it diminishes as prison populations expand; and
3) The overwhelming and undisputed negative side effects of incarceration far outweigh its potential, unproven
benefits.

These studies have led former pro-incarceration advocates to change their positions. James Q. Wilson now
concedes that we have reached a tipping point of “diminishing returns” on our investment in prisons.28 According
to Wilson, judges have always been tough on violent offenders and have incarcerated them for relatively long
periods. However, as states expanded incarceration, they
dipped “deeper into the bucket of persons eligible for prison, dredging up people with shorter and shorter criminal
records.”29 Increasing the proportion of convicted criminals sent to prison, like lengthening time served beyond
some point, has produced diminishing marginal returns
in crime reductions. Similarly, former incarceration advocates such as John DiIulio and former U.S. Attorney
General Edwin Meese are calling for a repeal of mandatory minimum sentencing and challenging the wisdom of
a massive imprisonment policy.30

THE NEGATIVE SIDE EFFECTS OF INCARCERATION
Incarceration may not have had much impact on
crime, but it has had numerous unintended consequences, ranging from racial injustice and damage to families
and children to worsening public health, civic disengagement, and even increases in crime.
Bruce Western demonstrates the extraordinarily disparate impact of imprisonment on young black males
compared to any other subgroup of society. For example,
he shows that nearly one-half of all young black males
who have not finished high school are behind bars, an
incarceration rate that is six times higher than for white
male dropouts. He then shows how incarceration damages the lifetime earnings, labor market participation, and
marriage prospects for those who have been to prison
and concludes that the U.S. prison system exacerbates
and sustains racial inequality.31
British penologists Joseph Murray and David
Farrington have analyzed data sets about child development from three nations and found that parental incarceration contributes to higher rates of delinquency, mental illness, and drug abuse, and reduces levels of school
success and later employment among their children.

Franklin E. Zimring and Gordon Hawkins. Incapacitation: Penal Confinement and the Restraint of Crime. New York: Oxford UP, 1995. 101; Thomas B. Marvell and Carlisle
E. Moody Jr. “Prison Population Growth and Crime Reduction.” Journal of Quantitative Criminology 10.2 (1994): 109-40, similarly found just a small reduction in crime resulting
from national prison expansion.
26
William Spelman. “The Limited Importance of Prison Expansion.” The Crime Drop in America, Revised edition. Ed. Alfred Blumstein. New York: Cambridge UP, 2006.
97-129. Spelman based this estimate on the amount of crime recently incarcerated felons told interviewers they had committed the year before their arrest. Spelman then
assumed that incarcerating and incapacitating those individuals averted the same amount of crime he assumed would be committed subsequently.
27
Bruce Western. Punishment and Inequality in America. New York: Russell Sage, 2006; James P. Lynch and William J. Sabol. “Did Getting Tough on Crime Pay?” Crime
Policy Report No. 1. Washington, DC: Urban Institute, 1997. http://www.urban.org/publications/307337.html; Don Stemen. Reconsidering Incarceration: New Directions for
Reducing Crime. New York: Vera Institute of Justice, 2007.
28
James Q. Wilson. “Crime and Public Policy.” Crime. James Q. Wilson and Joan Petersilia. Oakland, CA: ICS Press, 1995: 489-507.
29
Wilson, op. cit., 501.
30
Jacob Sullum. “Prison Conversion: After studying non-violent drug offenders, a criminologist who once said, “Let ‘Em Rot” now says “Let ‘Em Go.” Reason Aug./Sept. 1999.
31
Bruce Western, op. cit. See also, Raphael, op. cit.
25

9

Why and How to Reduce America’s Prison Population

Their comparative analysis of data from high-imprisonment and low-imprisonment nations reveals that the
negative effects of incarceration are most pronounced
when a nation’s incarceration rate is high, as it is in the
United States.32
Epidemiologist James Thomas has analyzed incarceration rates in North Carolina, and shows that high rates of
incarceration are associated with increased rates of teenage single-parent births and the risk of sexually-transmitted diseases among women. Other analyses have
shown that the high rate of HIV among black women can
be attributed to incarceration rates of black men.33
There are an estimated five million Americans who
are ineligible to vote because of felony convictions.
About half of these are African-Americans; consequently,
black communities’ have lost a significant amount of political power. Christopher Uggen and Jeffrey Manza have
demonstrated that the loss of voting rights by felons has

changed election outcomes across the nation, including
in the two most recent presidential elections, weakened
the political influence of minority communities, and detracted from civic participation by large classes of people
of color.34
Clearly, prison terms have a residual impact on the
families and communities of the imprisoned. Enduring
years of separation from family and community—deprived of material possessions, subjected to high levels
of noise and artificial light, crowded conditions and/or
solitary confinement, devoid of privacy, with reduced options, arbitrary control, disrespect, and economic exploitation—is maddening and profoundly deleterious. Anger,
frustration, and a burning sense of injustice, coupled
with the crippling processes inherent in imprisonment,
significantly reduce the likelihood that prisoners are
able to pursue a viable, relatively conventional life after
release.

32
Joseph Murray and David Farrington. “Effects of parental imprisonment on children.” Crime & Justice: A Review of Research. Forthcoming; and Joseph Murray, Carl-Gunnar
Janson, and David Farrington. “Crime in Adult Offspring of Prisoners: A Cross-National Comparison of Two Longitudinal Samples.” Criminal Justice & Behavior 34 (2007):
133-49.
33
James Thomas and Elizabeth Torrone. “Incarceration as Forced Migration: Effects on Selected Community Health Outcomes.” American Journal of Public Health 96 (2006):
1762–65; Rucker C. Johnson and Steven Raphael. “The Effects of Male Incarceration on Dynamics of AIDS Infection Rates Among African-American Women and Men.”
Unpublished paper presented to the incarceration study group of the Russell Sage Foundation (July 2006).
34
Christopher Uggen and Jeff Manza. Locked Out: Felon Disenfranchisement and American Democracy. New York: Oxford UP, 2006.

10

II
Why has there
been such a growth
in imprisonment?
The incarceration
explosion has been
based on a series
of ideas that were
widely accepted and
broadly popular at
the time, but have
turned out to be
erroneous.

Three Key Myths
about Crime and Incarceration
The research summarized above
shows that mass incarceration has a limited potential for reducing crime and has
collateral consequences that outweigh its
benefits. Why then has there been such a
growth in imprisonment? The incarceration explosion has been based on a series
of ideas that were widely accepted once,
but have turned out to be erroneous. The
three main myths are:
1. There are “career criminals” we can
identify and whose imprisonment
will reduce crime;
2. Tougher penalties are needed to
protect the public from “dangerous”
criminals; and
3. Tougher penalties will deter criminals.
Because these ideas have been so
central we discuss them in some detail.

Myth #1: There are “career criminals”
we can identify and whose imprisonment will reduce crime.
One of the primary justifications for
lengthy sentences is that we can identify
the “career criminals” or “violent predators”
who commit most of the serious crime
and who are not deterred or rehabilitated
by short sentences or alternative punishments to incarceration. These people, it is
argued, must be “incapacitated” for long

periods of time to reduce crime significantly. However, scientific efforts to develop methods for identifying career criminals have failed.
In 1978, a National Academy of Science
study by a panel of prestigious criminologists concluded that while most people
have a relatively brief and modest criminal career, a small group persists in crime.
However, the panel was unable to produce
any valid methods for identifying the persistent criminals at the beginning of their
criminal careers.35
Shortly thereafter, a group of RAND
researchers thought they had found a way
to identify persons who would eventually
become repeat offenders. They discovered
a group of “high-rate offenders” among
samples of males imprisoned for robbery
and burglary in three states (California,
Texas and Michigan).36 Significantly, the
vast majority of the surveyed prisoners reported committing very few if any crimes
the year before they were arrested (five or
less). But the so-called “high-rate” offenders, who comprised 10% of the samples,
claimed they had committed 4-5 robberies and burglaries per week in the preceding year. The researchers developed a
profiling scale that distinguished the highrate offenders, whom they labeled “violent
career criminals”, from the others. Peter
Greenwood claimed that if the people so

35
Jacqueline Cohen. “Research on Criminal Career: Individual Frequency Rates and Offense Seriousness.” Criminal Careers and
“Career Criminals. Alfred Blumstein et. al., eds. Washington, DC: National Academy Press, 1986.
36
Peter Greenwood and Alan Abrahamse, op. cit.; Jan M. Chaiken and Marcia Chaiken. Varieties of Criminal Behavior. Santa
Monica, CA: RAND, 1982.

11

identified were incarcerated for at least eight years, crime
would decline by 20%. Because the lower-rate offenders
could be released after serving reduced sentences, prison
expansion could be avoided.
These claims were widely disseminated by the U.S.
Department of Justice and used by elected officials to
justify sentencing reforms such as mandatory minimums,
truth in sentencing, and the abolition of parole boards.
However, when Greenwood and his associate Susan
Turner later followed a group of released prisoners identified as high-rate offenders by their profiles, they found
that these individuals had not committed crimes of the
type and at the rate expected.37
More recent research, nearly twenty years after the
first studies on the topic, continues to discredit the claim
that career criminals can be identified early by a profiling
system. John Laub and Robert Sampson have re-examined data from Sheldon and Eleanor Glueck’s 1930s seminal publication following the life careers of 500 Boston
delinquents.38 Although the vast majority desisted from
crime after the age of 25, a small minority persisted in
committing crime into their later years. Using all available
criteria, Laub and Samson could not distinguish these
“persisters” at the beginning of their delinquent careers
from the others who had followed the normal pattern of
criminal involvement in adolescence and desistance after
their early twenties.
Laub and Sampson were able to find a different set
of predictive factors, none of which could be observed
when the young people first committed crimes. Instead,
they found there were major “turning points” in a person’s
life—such as getting and holding a good job, enlisting in
the military, marrying, and establishing contacts with conventional institutions and groups—rather than personality characteristics or early childhood experiences that dis-

tinguished the careers of “desisters” from “persisters”. Laub
and Sampson also found that delinquents who had been
incarcerated were more likely to commit crimes later in
life than those who had been sentenced to probation or
local jail time. The implication was that imprisonment itself can encourage criminality.
In a subsequent study, Michael E. Ezell and Lawrence
E. Cohen published their analysis of the criminal careers of
three cohorts of “serious chronic offenders” released from
California youth prisons in the 1980s and 1990s. Although
the offenders varied in the amount of crime they committed after their release from prison, all “aged out” of crime.39
Similar to the other studies, Ezell and Cohen could not find
any background characteristics that reliably distinguished
those with different post-release criminal trajectories.
Other researchers have employed mixes of psychological test scores and behavioral measures to predict serious violent recidivism or involvement in new sex crimes.40
These methods predict recidivist crime better than random guessing, but they are also inaccurate.
Though improvements in statistical methods for
predicting violent recidivism have produced modest
gains in accuracy, they rely on data that would normally
not be available in a typical criminal justice application.41
Criminologists have been unable to develop practical and reliable methods to select those who will become career criminals.42 Attempts to incarcerate based
on any predictive criteria will inevitably end up incarcerating a large number of people who do not persist in
serious crime. As advocated later in the report, sentencing should not be based on what we think a person will
do but rather on what they have done and in proportion
to the seriousness of the crime.

Peter Greenwood and Susan Turner. Selective Incapacitation Revisited: Why High-Rate Offenders Are Hard to Predict. Santa Monica, CA: RAND, 1987. Christy A. Visher
also found methodological weaknesses in the study; see “The Rand Inmate Survey: A Reanalysis.” Criminal Careers and “Career Criminals,” Vol. 2. Alfred Blumstein et. al.,
eds. Washington, DC: National Academy Press, 1986.
38
Sheldon and Eleanor T. Glueck. Unraveling Juvenile Delinquency. New York: The Commonwealth Fund, 1950. John Laub and Robert Sampson continued the follow up
study of this cohort into the members’ 70’s. John Laub and Robert Sampson. Shared Beginnings, Divergent Lives: Delinquent Boys to Age 70. Cambridge, MA: Harvard UP,
2006.
39
Mike E. Ezell and Lawrence E. Cohen. Desisting from Crime: Continuity and Change in Long-Term Crime Patterns of Serious Chronic Offenders. New York: Oxford UP,
2005. 163. This “aging out” of crime is a consistent finding of most research on criminal careers. See, for example, David Greenberg. “Delinquency and the Age Structure of
Society.” Contemporary Crises 1.2 (1977): 189-224; David Farrington. “Age and Crime.” Crime and Justice: An Annual Review of Research. Michael Tonry and Norval Morris,
eds. Chicago: University of Chicago Press, 1986: 189-250.
40
Erica Beechner-Monas and Edgar Garica-Rill. “Genetic Predictions of Future Dangerousness: Is There a Blueprint for Violence?” Law and Contemporary Problems 69
(2006): 301-41; Richard Wollert. “Low Base Rates Limit Expert Certainty When Current Actuarials Are Used to Identify Sexually Violent Predators: An Application of Bayes’
Theorem.” Psychology, Public Policy, and Law 12 (2006): 56-85; Robert A. Prentky et. al. “Sexually Violent Predators in the Courtroom: Science on Trial.” Psychology, Public
Policy, and Law 12 (2006): 357-93.
41
John Monahan. “The Future of Violence Risk Management.” The Future of Imprisonment. Ed. Michael Tonry. New York: Oxford UP, 2004. 237-63.
42
Kathleen Auerhahn. “Selective Incapacitation and the Problem of Prediction.” Criminology 37.4 (1999): 703-34; Jacqueline Cohen. “Incapacitation as a Strategy for Crime
Control: Possibilities and Pitfalls.”Crime and Justice: An Annual Review of Research 5 (1983): 1-84; Jacqueline Cohen and Jose A. Canela-Cacho. “Incarceration and Violent
Crime, 1965-1988.” Understanding and Preventing Violence: Consequence and Control, Vol. 4. Albert J. Reiss and Jeffrey A. Roth, eds. Washington, DC: National Academy
Press, 1994; Rudy A. Haapanen. Selective Incapacitation and the Serious Offender: A Longitudinal Study of Criminal Career Patterns. New York: Springer-Verlag, 1990.
37

12

Unlocking America

Table 4: Percent of Arrests Attributed to Released Prisoners
Arrests in 7 States

Number

Total Arrests in 1994-97

Percent
2,994,868

100

140,543

5

36,000

1

Total Arrests of Prisoners Released in 1994-97
Total Arrests of Released Prisoners for Violent Crimes
Source: Recidivism of Prisoners Released in 1994. US DOJ. Washington, DC: Bureau of Justice Statistics, 2002

Myth #2: Tougher penalties are needed to protect
the public from “dangerous” criminals.
We can’t seriously address our crime problem until
repeat offenders learn that they will get no more breaks
from the criminal justice system. … When incarcerations
increase, serious crimes go down.
—Andrew Thomas, County Attorney, Maricopa County, Arizona

43

The failure of efforts to develop methods of accurately identifying the small number of offenders who do
commit particularly horrendous crimes after serving their
sentences fueled demands for longer sentences across
the board.44 The logic of this argument was that if we can’t
single out the truly dangerous, we will assume that anyone with two or three convictions for a relatively wide
range of offenses is a dangerous habitual criminal, and
keep them all in prison for an extremely long time. On the
basis of this reasoning, a number of states adopted mandatory sentencing, truth in sentencing and in some states
“three strikes” laws, all of which extend prison sentences.

These laws have done little to reduce crime. Few
convicted persons have the requisite number of previous felony convictions to qualify for the enhanced sentences.45 This is because rates of return to serious crime
on the part of those released from prison are not high.
Just 1.2% of those who served time for homicide and
were released in 1994 were rearrested for a new homicide
within three years of release, and just 2.5% of released
rapists were arrested for another rape. Sex offenders were
less likely than non-sex-offenders to be rearrested for
any offense.46 Their rates of re-arrest for a new sex offense
were only 5.3%. A substantial percentage (over 60%) of
released prisoners are eventually rearrested, but mostly
for drug offenses or violations of parole regulations. Many
of those eligible for sentence enhancements under these
“habitual offender” statutes are at an age when they are
reducing their involvement in crime, or abandoning it altogether.
The U.S. Department of Justice conducted a major
study of criminal involvement of prisoners who had been
released in 1994. It found that only 5% of the 3 million ar-

Table 5: Chances of Being Caught
Indicator
Victimizations

Number

% of Total
25,036,030

100

9,721,205

39

13,699,254

55

4,642,803

19

State and Federal Felony Convictions

983,823

4

Convictions Resulting In Prison or Jail

680,000

3

Reported to Police
Arrests
Less Drug, Alcohol, and Other Victimless Crimes

Source: U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics, Prison Statistics. Online. Available: http://www.ojp.usdoj.gov/bjs/keyfacts.
htm. Accessed: August 1, 2006.

Christian Richardson. “Thomas: Probation Out in Major Felony Cases.” East Valley Tribune 29 Nov. 2006.
Jonathan Simon. “Reversal of Fortune: The Resurgence of Individual Risk Assessment in Criminal Justice.” Annual Review of Law and Social Science 1 (2005): 397-421;
Bernard E. Harcourt. Against Prediction: Profiling, Policing, and Punishing in an Actuarial Age. Chicago: University of Chicago Press, 2007.
45
Franklin E. Zimring, Gordon Hawkins, and Sam Kamin. Punishment and Democracy: Three Strikes and You’re Out in California. New York: Oxford UP, 2001.
46
A meta-analysis of 61 studies also concluded that sex offenders have low rates of recidivism. See R. Karl Hanson and Monique T. Bussierre. “Predicting Relapse: A MetaAnalysis of Sexual Offender Recidivism Studies.” Journal of Consulting and Clinical Psychology 66 (1998): 348-62; and R. Karl Hanson et. al. “Sex Offender Recidivism: What
We Know and What We Need to Know.” Annals of the New York Academy of Science, Sexually Coercive Behavior: Understanding and Management. Robert A. Prentky et. al.,
eds. (2003): 154-.
43
44

13

Why and How to Reduce America’s Prison Population

rests made in seven states between 1994 and 1997 were of
recently released prisoners (Table 4).47 California’s “three
strikes” law has had a number of evaluations; almost all
found that it failed to reduce crime.48
These studies make clear that, while many people
who are released from prison end up back behind bars,
they are but a fraction of the overall crime problem.
Lengthening their sentences, as a means of dealing with
crime will at best have only marginal impact.

Myth #3: Tougher penalties will deter criminals
The hope that the presence of brutal prisons will deter law violators is as old as the invention of the prison itself. Contemporary supporters of longer sentences argue
that longer and harsher punishments are necessary to
deter crime by making criminals think twice before committing another crime. This notion simplifies and distorts
the dynamics of criminal behavior. When most people
commit a crime, they correctly believe that they will not
be caught. According to NCVS surveys, about 60% of all
crimes are not reported to police. The FBI reports that 4.6
million arrests were made for property or violent crimes,
a number representing only 19% of the reported victimizations. State and federal courts dealing with serious
felony crimes produce about one million convictions—of
which 68% result in a prison or jail sentence. Therefore,
as illustrated in Table 5, the immediate chances of being
arrested, convicted, and sentenced to jail or prison are
quite small.
Further, different types of crimes vary markedly in
their potential for being deterred and people differ in
their susceptibility to being deterred by the threat of
punishment. The population roughly divides into three
groups: 1) those who, because of personal values and so-

cial commitments, would not knowingly commit crimes
under any circumstances; 2) those who are deterred
when they believe that there is some likelihood they will
be caught and punished; and 3) those who are not deterred. The type of crime most likely to be reported and
to result in arrest and conviction tends to be committed
by those least likely to be deterred—generally young
males excluded from the conventional pathways to success, many of whom already have been severely punished
by the juvenile justice system early in their lives and are
unlikely to be deterred in the future. The vast majority
of these youth desist from crime after their twenties for
reasons unrelated to any penalties the state imposes on
them. Ironically, those who are deterred when punishment is seen as likely—polluters, price fixers, slum lords,
inside traders, stock swindlers, workplace safety violators,
and other “white-collar” criminals—receive relatively lenient punishment when caught.
At the turn of the 19th century reformers realized
that brutal prisons embitter prisoners rather than reform
them. Yet this persistent faith that prisoners can be discouraged from returning to crime by subjecting them
to harsh penalties, or that the population at large can be
deterred more effectively with severe penalties than with
milder ones, has never had empirical support. Decades of
research on capital punishment have failed to produce
compelling evidence that it prevents homicide more
effectively than long prison sentences.49 Community
penalties, it has been shown, are at least as effective in
discouraging return to crime as institutional penalties.50
Rigorous prison conditions substantially increase recidivism.51 Evaluations show that boot camps and “scared
straight” programs either have no effect on recidivism or
increase it.52 “Tough” sanctions are popular, but they do
not reduce crime.

Recidivism of Prisoners Released in 1994. US DOJ. Washington, DC: Bureau of Justice Statistics, 2002.
John L. Worrall. “The Effect of Three-Strikes Legislation on Serious Crime in California.” Journal of Criminal Justice 32 (2004): 283-96.
Sarah T. Dike. Capital Punishment in the United States. Hackensack, NJ: National Council on Crime and Delinquency, 1982; Richard A. Berk. “New Claims about
Executions and General Deterrence: Deja Vu All Over Again?” Journal of Empirical Legal Studies 27 (1998): 209-19.
50
Joan Petersilia and Susan Turner. Prison versus Probation in California: Implications for Crime and Offender Recidivism. Santa Monica, CA: RAND, 1986.
51
M. Keith Chen and Jesse M. Shapiro. 4 Dec. 2006. “Does Prison Harden Inmates? A Discontinuity-Based Approach.” Cowles Foundation, Harvard University http://home.
uchicago.edu/c~jmshapir/prison120406.pdf/
52
Doris Layton MacKenzie and James O. Finckenauer. Scared Straight: Delinquency and the Panacea Phenomenon. Englewood Cliffs, NJ: Prentice Hall, 1982; Doris Layton
MacKenzie. What Works in Corrections; Reducing the Criminal Activities of Offenders and Delinquents. New York: Cambridge UP, 2006; Mark W. Lipsey. “What Do We Learn
from 400 Research Studies on the Effectiveness of Treatment with Juvenile Delinquents?” What Works: Reducing Reoffending. Ed. J. McGuri. New York: Wiley, 1995.
47
48
49

14

III
The danger of
relying on treatment
and programs to
solve America’s
imprisonment crisis is
that when recidivism
isn’t reduced,
imprisonment will be
regarded as the only
viable answer to the
crime problem.

The Limits of Prison-Based
Rehabilitative and Treatment Programs
There is a growing belief that prison
populations can be reduced simply by
expanding rehabilitative and treatment
programs within our prisons. This position
is based largely on new studies showing
that treatment or rehabilitation services
can reduce recidivism and thus reduce
the number of persons being re-admitted to prison after release. Increasingly,
completion of rehabilitative or addiction
treatment programs can and do result in
a reduced period of confinement.
Our concern is that we delude ourselves that treatment programs hold the
key to reducing the prison population.
Prison populations did not increase in the
1970s because programs disappeared,

causing crime and recidivism rates to increase. Rather—and it is important to be
clear about this—prison populations increased because laws were passed that
increased the number of people sent to
prison and their length of stay.
Even if treatment programs were effective in reducing recidivism, recidivism
is not the primary cause of the explosion
in the prison population, as we have argued above. Evidence suggests, however,
that such programs are in fact not particularly effective in reducing even the
portion of prison overcrowding caused
by recidivism. The most recent probation
and parole success rates and the reasons
for not completing probation or parole

Table 6: Probation and Parole Success Rates—1995-2003
OUTCOME MEASURES

PROBATION

PAROLE

1995

62%

45%

2000

60%

43%

2003

59%

47%

Re-incarcerated

16%

38%

New Conviction and Sentence

5%

11%

Revocation

7%

26%

Other

4%

1%

Absconded

4%

9%

Other

22%

6%

SUCCESSFUL COMPLETIONS

REASON FOR FAILURES

Source: Probation and Parole in the United States, 2003. US DOJ. Washington, DC: Bureau of Justice Statistics, 2004.

15

Table 7: 1983 and 1994 Prisoner
Three Year Follow-Up Recidivism Rates
Recidivism
Measure

1983 Prison
Releases

1994 Prison
Releases

Re-Arrested

63%

69%

Re-Convicted

47%

47%

Re-Imprisoned

41%

40-52%*

*52% readmission to prison with California prisoners and 40% without California
prisoners included.
Sources: Recidivism of Prisoners Released in 1994. US DOJ. Washington, DC:
Bureau of Justice Statistics, 2002; Recidivism of Prisoners Released in 1983.
US DOJ. Washington, DC: Bureau of Justice Statistics, 1989.

supervision have not changed since the U.S. Department
of Justice began reporting them in 1995 (Table 6). The
same holds true for recidivism rates of prisoners released
from prison (Table 7). The Bureau of Justice Statistics
national recidivism rates of prisoners after release have
not changed between 1983 and 1994. Figure 4 shows
recidivism rates for the states of Washington and New
York, which have been recording their three-year return
prison rates as early as 1985. Thus despite a number of reforms that have been designed to increase the availability
of program services, recidivism rates have remained unchanged.53
A growing literature on “what works” in correctional
programming has found that many programs have no
impact on recidivism rates. A recent “meta-analysis” of
treatment programs reviewed 291 evaluations of adult
offender treatment programs, both in-prison and in-community, conducted in the United States and other Englishspeaking nations.54 It reports that 42% of the evaluated
programs, including jail diversion programs, domestic violence programs, faith-based, psychotherapy or behavior
therapy for sex offenders, boot camps, electronic monitoring, and restorative justice programs had no impact on
recidivism.
Of the 167 effective programs, only one-fourth were
prison-based treatment programs. Of these programs,

the reduction in recidivism rates generally ranged from
4% to 10%. These recidivism reduction results are similar
to those found in other major evaluations of “what works”
(or “doesn’t work”). Notably, many other programs do not
reduce recidivism and may actually increase rates of failure.55
Treatment and rehabilitative programs tend to be
most effective when they are disassociated from government coercion.56 Someone who doesn’t want to be
rehabilitated is not a promising candidate for being rehabilitated. Requiring someone to sit through a program
designed to deal with dependence on alcohol or drugs
may lead to resentment and shammed participation. It
is not likely to bring about the inner transformation that
will end involvement in crime. The association of help
with punishment that occurs when courts require treatment is likely to discourage those who could benefit from
programs to seek them out on their own. The presence of
substantial numbers of reluctant participants may undermine the quality of the programs, reducing their ability to
help those who want to be helped. Those who administer programs are likely to become demoralized if they are
forced to accept “clients” sent to them by a court, rather
than working with those who seek help on their own.
Even under the most optimistic assumptions, the effect of in-prison rehabilitative programs is to reduce failure rates by 10% or so. If all correctional programs were
to achieve this level of success, they would reduce failure
rates for about one-third of those confined from about
40% to about 30%. A 10% reduction in return-to-prison
rate for one third of the population translates to an overall
system rate reduction in recidivism of 3%. This is not large
enough to substantially reduce a prison population.
The inability of rehabilitation programs to stem prison
expansion can be seen in several major states. The Florida
Department of Corrections reported that it offered a wide
array of treatment, education and vocational programs to
over 6,500 prisoners released in 1995 and 1996. For those
who completed the programs, a recidivism rate reduction of 6-10% was achieved. But the estimated number
of prisoners thus not returning to prison because of involvement in the programs was about 400 or two percent
of the more than 18,000 persons admitted to prison in

53
See “State of New York Department of Correctional Services” (undated), 2001 Releases: Three Year Post-Release Follow-Up. Albany, New York; and the Washington State
Department of Corrections Web site at: http://www.doc.wa.gov/BudgetAndResearch/ResearchData/Recidivism20.pdf/
54
Steve Aos, Marna Miller, and Elizabeth Drake. Evidence-Based Adult Corrections Programs: What Works and What Does Not. Olympia, WA: Washington State Institute for
Public Policy, 2006. Only 20% of the evaluations used random assignment procedures to create treatment and control groups. The other studies used “matching techniques”
to approximate the attributes of the “treatment” group, which is less rigorous.
55
Donald A. Andrews. “Principles of Effective Correctional Programs.” Compendium 2000 on Effective Correctional Programming. Laurence L. Motiuk and Ralph C. Serin, eds.
Ottawa: Correctional Service Canada, 2001; Larry Sherman et. al. Preventing Crime: What Works, What Doesn’t, What’s Promising. Washington, DC: National Institute of
Justice, 2006.
56
Ojmarrh Mitchell, David B. Wilson, and Doris L. MacKenzie, “The Effectiveness of Incarceration-Based Drug Treatment on Criminal Behavior,” which is a meta-analysis of
in-prison treatment programs for drugs.

16

Unlocking America

the same year. Indeed, while these
programs operate, both prison admissions and the prison population
have continued to escalate.57
In California, a recent study
by the state’s Inspector General’s
Office concluded that California’s
$1 billion investment in drug treatment for prisoners since 1989 has
been “a complete waste of money,”
and has failed to reduce recidivism
rates. In fact a study by UCLA found
that the state’s two largest in-prison
programs produced higher recidivism rates for inmates who participated in the programs as opposed
to those who did not receive treatment.58 Some of the reasons cited
for the lack of impact are improper
program design and operations.
With respect to treatment in
lieu of incarceration, we agree that large numbers of persons charged with felony crimes should be sentenced to
alternatives to state prison that also provide treatment
services and education programs. But numerous studies of prison diversion programs show that these, too,
may not reduce prison admissions. Too often, alternative
programs have been reserved for people who would not
otherwise have been sentenced to prison.59 In some instances, diversion programs actually increase prison populations. This can occur when people who plead guilty
to be eligible for treatment fail to complete the program
and end up being sent to prison. In many of these cases,
they would have had their charges dropped if they had
not pled guilty or been given a county jail sentence or a
shorter prison sentence than the one they received after
failing the treatment program.60
Ultimately, the purpose of treatment and rehabilitation—and the criteria for assessing their usefulness and
efficacy—should not be a major reduction in recidivism,

Figure 4: Three-Year Recidivism Rates

much less reducing the size of the prison population.
These are unreasonable burdens to place on programs
that are, by their very nature, limited.
There is no question that providing meaningful
work, education, and self-development programs to
prisoners promotes more human and safer prisons. And
a growing body of research, as noted below, suggests
that prisoners who seriously take advantage of well-administered rehabilitative services and complete the programs are more likely to succeed in achieving satisfying
conventional lives after prison than persons who do not
receive these services. But expanding treatment services
has not been shown to be an effective means of reducing prison populations. Indeed, relying on treatment and
programs to solve America’s imprisonment crisis carries
the risk that, when they fail to reduce recidivism, imprisonment will be regarded as the only viable answer to the
crime problem.61 Treatment programs are necessary and
humane, but they are not answers to the crisis of prison
overpopulation.

Analysis of the Impact of Inmate Programs on Recidivism. Tallahassee: Florida Department of Corrections, 2001 http://www.dc.state.fl.us/pub/recidivismprog/index.html/.
Matthew L. Cate. “Special Review into In-Prison Substance Abuse Treatment Programs Managed by the California Department of Corrections and Rehabilitation.”
Sacramento: California Bureau of Audits and Investigations, 2007.
59
Larry Sherman et. al., op. cit.
60
Joan Petersilia. “A Decade of Experimenting with Intermediate Sanctions: What Have We Learned?” Corrections Management Quarterly 3.3 (1999): 19-27.
61
In 1973, New York Governor Nelson Rockefeller proposed the infamous Rockefeller drug laws, imposing life sentences for possessing or selling drugs, after concluding that
treatment programs for heroin users were not working. Other states followed suit with similarly long sentences for drug-law violators. These laws are widely seen as failures
because they imprison low-level “mules” or street-level dealers for very long stretches, without capturing major operators or substantially reducing drug use. It is only because
so much faith had been placed in rehabilitation that the publication of Robert Martinson’s conclusion that most treatment programs had not been shown to be effective could
have been construed, 30 years ago, as providing support for prison expansion. Robert Martinson. “What Works? Questions and Answers about Prison Reform.” The Public
Interest 35 (1974): 22-54.
57
58

17

Decarceration,
Cost Savings, and Public Safety
The overuse of incarceration, along
with the mistaken justifications that have
supported this policy, have corrupted and
compromised our criminal justice policies
and paralyzed efforts to reform them. The
net result is an expensive system that relies much too heavily on imprisonment, is
increasingly ineffective, and diverts large
sums of taxpayers’ money from more effective crime control strategies. Even
some of the leading criminologists who
lobbied for putting more people in prison
are now advocating that we invest some
of the tens of billions we are now spending on imprisonment in more productive
crime-reduction practices.
The impact of a $1 million investment
in ... cash and other incentives to
disadvantaged students to graduate
from high school would result in a
reduction of 258 crimes per year, and
parent training therapy for families with
young “acting out” children 160 crimes
per year, compared to a reduction of
60 crimes a year through building and
operating prisons.62

But to make such investments we
must reduce the system that is draining
our resources. The facts indicate that placing large numbers of people in prison and

jail or on probation and parole for long
periods of time is not an effective crime
control strategy. One study found that after imprisonment exceeded a certain tipping point, it became counter-productive.
When too many men are removed from a
community, family and social life are destabilized, leading to higher crime rates.63
Making matters worse, we have
squandered lean government budgets
on corrections (responding to the symptom) instead of funding programs that
shepherd dislocated young people onto
conventional paths (responding to the
cause). While we build prisons, we do not
increase funding for academic and technical education, job training, healthcare, affordable housing, or other social services
that assist people in getting a foot in the
door to a better life.
The loss of manufacturing jobs in recent decades has reduced the opportunities for non-white, inner-city youth to get
their feet in that door.64 Prison expansion,
by stigmatizing such a large number of
black and Latino male youth, has placed
an additional obstacle in their path. It interferes with family formation, strains existing family relationships, and disrupts
careers.65 Because employers commonly
discriminate against minorities and also

IV
The facts indicate
that placing so many
people in prison and
jail or on probation
and parole for long
periods of time is not
an effective crime
control strategy.

Peter W. Greenwood et. al. Diverting Children from a Life of Crime. Santa Monica, CA: RAND, 1998.
Dina R. Rose and Todd R. Clear. “The Problem with ‘Addition by Subtraction’: The Prison-Crime Relationship in Low-Income
Communities.” Invisible Punishment: The Collateral Consequences of Mass Imprisonment; Marc Mauer and Meda ChesneyLind, eds. New York: The New Press, 2003. 181-94;Todd R. Clear. Imprisoning Communities: How Mass Incarceration Makes
Disadvantaged Places Worse. New York: Oxford UP, 2007.
64
John M. Hagedorn. People and Folks: Gangs, Crime, and the Underclass in a Rustbelt City. Chicago: Lake View Press, 1988;
Loic Wacquant. “When Ghetto and Prison Meet and Mesh.” Punishment and Society 3.1 (2001): 95-134; Steven Raphael,
op. cit.; Bruce Western, op. cit.; Thomas J. Sugrue, The Origins of the Urban Crisis: Race and Inequality in Postwar Detroit.
Princeton, NJ: Princeton UP, 1996.
65
John Hagan and Ronit Dinovitzer. “Collateral Consequences of Imprisonment for Children, Communities, and Prisoners.”
Crime and Justice: A Review of Research 26 (1999): 121-62; Bruce Western. Punishment and Inequality in America. New York:
Russell Sage, 2006.
62
63

18

Unlocking America

against ex-convicts, our current crime-control strategy
dooms many people who start life with major disadvantages. The long-term association of minority status with
crime in the public mind is strengthened when a highly
disproportionate number of blacks and Latinos are locked
up.
In the long run, these stereotypes reinforce discriminatory residential and employment patterns. Prospective
employers, faced with a black male job applicant, are now
more likely than in the past to suspect that the applicant
has a criminal record, and decline to offer a job.66 High
rates of imprisonment may thus reduce legitimate stakes
in conformity and eliminate lawful alternatives to crime,
not just for those who are sent to prison but also for those
who are not.
Careful analysis of variations in states’ crime and incarceration rates reveals a consistent relationship: states
with the lowest crime rates also have the lowest incarceration rates, and this is not primarily a result of incarceration reducing crime. Put differently, if incarceration
were the key to a safer society, cities and states with
exceptionally high incarceration rates (e.g., Baltimore,
Washington, D.C., Louisiana, Texas, and Oklahoma) would
be the safest—not the most dangerous—places to live.
What makes a place safe are social and economic factors

that deliver a high quality of life as measured by good
education, strong families, informal social controls, viable
networks, and opportunities for stable, meaningful, and
well-paid work.67
While incarceration does not make us safer, decarceration, as recommended in this report, would free up
the resources necessary to produce neighborhoods that
are good places to live, work, and play, making people
safe (and feel safe) from common forms of interpersonal
violence and theft. It lies beyond the scope of this document to specify a program for doing this. There are, however, steps that government, business, and community
organizations can take in preventing communities from
deteriorating and in helping them improve when they
are in trouble. The criminal justice system has a role to
play in this process. Businesses will be reluctant to invest
in communities where they will be at high risk for violence and theft. Neighbors will be reluctant to participate
in and take responsibility for community affairs when
the streets are unsafe. At the same time, it must be recognized that criminal justice agencies cannot do the job
alone. We focus here on criminal justice, but insist that
the contribution it makes must be informed by key facts
we have spelled out above if the result is to be progressive, humane, and effective.

66
John R. Lott Jr. “The Effect of Conviction on the Legitimate Income of Criminals,” Economic Letters 34 (1990): 381-85; Devah Pager. “The Mark of a Criminal Record.”
American Journal of Sociology 108 (2003): 937-75 and “Double Jeopardy: Race, Crime, and Getting a Job,” Wisconsin Law Review (2005): 617-60; Robert J. Sampson and
John H. Laub. Crime in the Making: Pathways and Turning Points Throughout Life. Cambridge, MA: Harvard UP, 1993; Joel Waldfogel. “The Effect of Criminal Conviction on
Income and the Trust Reposed in the Workmen.” Journal of Human Resources 29 (1994): 62-81 and “Does Conviction Have Permanent Effect on Income and Employment?”
International Review of Law and Economics 18 (1998): 25-40; Jeffrey Grogger. “The Effect of Arrests on the Employment and Earnings of Young Men.” The Quarterly Journal
of Economics 11 (1995): 51-71; Marc Mauer. Race to Incarcerate. New York: New Press, 1999; Richard B. Freeman and Jeffrey Fagan. “Crime and Work.” Crime and Justice:
A Review of Research 265 (1999): 225-90; Marc Mauer and Meda Chesney-Lind, eds. Invisible Punishment: The Collateral Consequences of Mass Imprisonment. New York:
New Press, 2002; Michael Tonry. Malign Neglect. New York: Oxford UP, 1995; Bruce Western. Punishment and Inequality in America. New York: Russell Sage, 2006.
67
Arnold S. Linksy and Murray A. Straus. Social Stress in the United States: Links to Regional Patterns in Crime and Illness. Dover, MA: Auburn House, 1986.

19

Recommendations
Our Orienting Idea: Punishment Should Fit the Crime.
We have argued that using imprisonment to reduce crime by deterring, incapacitating, or rehabilitating is of limited value,
and is now yielding diminishing returns.
What then should imprisonment be used
for? The size of the literature addressing
this subject is not matched by its success in
forging a consensus on the matter, and we
do not intend to settle all debates here. That
said, we think it useful to be clear about our
own guiding thoughts.
Imprisonment can legitimately satisfy
a social and personal need for retribution toward those who violate society’s laws. Most
contemporary philosophies of punishment
give a large role to retribution.
In addition to satisfying victims’ needs,
punishing lawbreakers according to what
they deserve can perform important social
functions. Punishment can promote social
solidarity, while failure to respond to crime
weakens commitment to social norms. At
the same time, excessive punishment can
exacerbate social tensions and widens divisions, reducing solidarity. It can corrode a
nation’s political culture, and obstruct efforts
to deal constructively with social problems,
including crime.
Retribution should not be used as an
excuse for mindless punitiveness as is the
case now. The essence of the retribution is to
punish people proportionately to what they
deserve, based on the crime they have committed. Excessive leniency and undeserved
harshness both violate the principle of proportionality. Failure to limit the severity of

punishment to what is deserved is unjust. It
alienates citizens from the government and
undercuts the effectiveness of law enforcement. When those who are punished are
disproportionately poor and members of
minority groups, it is inevitable that they will
believe that the law is being used to repress
them, rather than holding them accountable
for their crimes.
Public opinion polls have consistently
shown that substantial numbers of people
think that the courts are too lenient. These
sentiments cannot be taken at face value,
and should not be allowed to dictate sentencing policy unthinkingly. For example, the
vast majority of crimes are neither as serious
as the public believes them to be nor as heinous as the media portrays them. As shown
in Table 8, very few (about 15%) of the crimes
for which people are arrested are either violent or serious property crimes. Yet many of
these arrests are resulting in prison terms.
We have already noted that the costs of
these crimes to the public are only a fraction
of the costs of punishing those who are arrested and convicted. The volume of serious
crime attributed to released prisoners is also
much lower than is commonly believed.
In addition, many people do not fully
appreciate how harsh and disruptive any
form of imprisonment is because they have
never experienced the total loss of agency
and privacy that imprisonment entails. Most
prisoners experience monotonous routines,
medical neglect, physical danger, extreme
isolation, and a myriad of deprivations—all

V
Our resources
are misspent, our
punishments too
severe, our sentences
too long.
—Justice Anthony M. Kennedy68

68
Justice Anthony M. Kennedy. “Speech at the American Bar Association, Annual Meeting: An Address by Anthony M. Kennedy,
Associate Justice, Supreme Court of the United States.” American Bar Association. San Francisco, CA. 9 Aug. 2003.

20

Unlocking America

of which worsen the trauma of imprisonment.
When people are presented a fuller picture of the facts
of particular crime and the criminals’ characteristics, they
generally favor more moderate sanctions. Recent studies
show that “when nonviolent offenders are involved, there
is substantial support for intermediate sanctions and for restorative justice.”
We also note that problems can potentially arise when
prosecutors, defense counsel and judges want to modify a
sentence by taking into account other factors that are more
related to the goals of deterrence, incapacitation and rehabilitation. For example, should we allow different prison
sentences for two people who have committed the same
crime but have different prior records or have been assessed
to present a lower risk to public safety (male versus female,
older versus younger, drug addiction versus none)? Similarly,
one could argue that someone who has committed only a
very minor crime but has been assessed as a “high risk” to
commit a serious crime if released and hence in need of being deterred, incapacitated or rehabilitated should be incarcerated as opposed to someone convicted of a more serious
crime but poses no such threat to public safety.
The prestigious American Law Institute’s recently issued
Model Penal Code endorses a concept of “limited retribution”,
which allows for the introduction of rehabilitative, deterrence
and incapacitation factors which can influence sentences
within minimum and maximum limits based on retributive
considerations. This position has the merit of limiting the extent to which sentence lengths can be extended far beyond
what a defendant deserves. Prosecutors and judges currently take such considerations into account as they negotiate
pleas and set sentences.
Nevertheless we are concerned about the potential
for injustice and discrimination associated with this practice.
When decisions made as to whether someone should be imprisoned or for how long on the basis of what crimes one
might commit if released, or on the basis of a person’s needs
for rehabilitation, they will often be incorrect. Subjecting
people who will not commit serious new crimes to prison
sentences (or to longer sentences) simply on the basis of
predictions that are false is simply unfair.
We are also concerned determinations of dangerousness or in need of treatment are likely to be skewed by racial and class biases. Racial stereotypes sometimes operate
unconsciously and can influence perceptions of dangerousness even on the part of decision-makers who harbor no
conscious prejudices.
Minority offenders’ personal circumstances may make
them appear to some judges as unlikely prospects for rehabilitation. Those who can pay for private drug or mental
health treatment, provide restitution in large amounts to vic-

21

Table 8: Arrests by Type of Crime, 2005
Crime Type
Total Arrests

Number
14,094,186

Total Serious Violent

Percent
100%

603,503

4%

14,062

0%

25,528

0%

Robbery

114,616

1%

Aggravated assault

449,297

3%

1,609,327

11%

298,835

2%

1,146,696

8%

147,459

1%

16,337

0%

1,846,351

13%

Alcohol/Liquor

2,525,924

18%

Driving under the influence

1,371,919

10%

Liquor laws

597,838

4%

Drunkenness

556,167

4%

All Other Crimes

10,035,005

53%

Other assaults

1,301,392

9%

Forgery and counterfeiting

118,455

1%

Fraud

321,521

2%

18,970

0%

Murder and non-negligent
manslaughter
Forcible rape

Total Serious Property
Burglary
Larceny theft
Motor vehicle theft
Arson
Drug abuse violations

Embezzlement
Stolen property; buying,
receiving, possessing
Vandalism

133,856

1%

279,562

2%

Weapons

193,469

1%

Prostitution

84,891

1%

Non-rape sex offenses

91,625

1%

Gambling

11,180

0%

Offenses against the family

129,128

1%

Disorderly conduct

678,231

5%

33,227

0%

3,863,785

27%

3,764

0%

Curfew/loitering

140,835

1%

Runaways

108,954

1%

Vagrancy
All other offenses
Suspicion

Source: UCR 2005, Federal Bureau of Investigation.
*Percentages do not add up to 100% due to rounding.

Why and How to Reduce America’s Prison Population

Table 9: Three Year Follow-Up
Rate of Re-Arrest of State Prisoners
Released in 1994, By Time Served in Prison
Time Served

Three Year Re-Arrest Rates
66.0%
64.8%
64.2%
65.4%
68.3%
62.6%
63.2%
54.0%

6 Months or Less
7-12 months
13-18 months
19-24 months
25-30 months
31-36 months
37-60 months
61 months or more
Source: Prison Statistics. US DOJ. 1 Aug. 2006 http://www.ojp.usdoj.gov/bjs/prisons.htm/

tims and communities, or attend educational and vocational
programs often unavailable to the poor are likely to receive
milder punishments than others who have committed exactly the same crimes.
This is especially so in the current political context,
which increasingly relies on the private sector to provide
correctional services. This means that middle-class criminals
with drug or mental health problems get help because they
can pay for it—and stay out of prison—while poor people
who cannot do so go to jail. By reducing the magnitudes
of sentences and restricting them within narrow limits,
such disparities are likely to be reduced. But we remain concerned that poor and non-white people might be likely to
receive harsher punishments within these limits.
For all of these reasons we oppose the practice of
imposing prison sentences “so that the defendant can be
rehabilitated,” or “to protect society” from “a dangerous”
person. On the other hand, retributivism does not require
that everyone who violates a given statute receive exactly
the same sentence. In determining the appropriate punishment in a particular case, some characteristics of the person
and the crime committed can legitimately be considered.
All violations of a particular criminal statute are not alike.
A sentencing system should be “flexible enough to permit
individualized sentences when warranted by mitigating or
aggravating factors not taken into account in the establishment of general sentencing practices.”
At the same time it should not be so unstructured as to
create “unwarranted sentencing disparities among defendants with similar records who have been found guilty of
similar criminal conduct.” Flexibility and equity in sentenc-

ing can thus be achieved by considering the criminal’s personal circumstances, the circumstances of the crime itself
(e.g. was it spontaneous or premeditated?) and criminal record to influence the sentence.
The implementation of these very general principles
is a matter that deserves careful attention by state legislators. It lies beyond the scope of this report to specify the details
of such a sentencing scheme. Nevertheless, we insist that the
presumption against imprisonment should hold firm in all
cases.
Given the excessive lengths of sentences now being
imposed on many defendants, a reconsideration of sentencing structures should begin with the premise that sentence
lengths should be reduced substantially. Furthermore, innovative methods of serving in-community sentences should
be explored. Especially promising are programs of victimoffender restoration and community reparation such as are
already operating in Europe, Australia and New Zealand.

FOUR RECOMMENDATIONS THAT WILL REDUCE
PRISON POPULATION
To reduce the prison population, we propose four recommendations. If implemented, they would dramatically
reduce current prison, parole, probation, and jail populations. All of these recommendations have been successfully
implemented in some fashion in several jurisdictions with
no adverse impact on crime rates.69 By lowering the prison
population, the associated costs—tens of billions of dollars
now spent in an ineffective war on crime—could either be
returned to tax payers or reinvested by the public and private sectors in those social, family, community, and economic

69
Reducing prison populations by shortening sentences and ceasing returning parole violators to prison could only occur over a span of several years and, therefore, there
would be little or no impact on crime rates due to influx of released prisoners. In several instances, prison populations have been reduced in a short period of time and there
was no resultant increase in crime rates. Similarly, decriminalization of drug crimes, even heroin and cocaine possession and use, has occurred with no increase in drug use
or other crimes.

22

Unlocking America

institutions that have more direct influence on crime rates.

RECOMMENDATION 1: Reduce time served in prison.
The fundamental and most powerful reform that must
occur if we are to have any hope of reversing the imprisonment binge is to reduce the severity of the sentences they
are given. For many prisoners now sent to prison, this would
mean probation or a short jail sentence; for others, it would
mean less time spent in prison, as well as less time on parole.
An in-depth examination of sentence lengths and time conserved is called for, it might begin with the presumption that
terms be cut back to what they were circa 1975 when the
imprisonment binge began.
This central recommendation is grounded in three facts:
1) many prisoners are now serving longer prison terms; 2) the
longer prison terms are not proportionate to the severity of
the crimes they were convicted of; and 3) the extension of
their length of incarceration has no major impact on their recidivism rate, or crime rates in general. As shown in Table 9
there is no association between length of stay and recidivism
rates. Coupled with the previous research finding that released prisoners account for a small percentage of all arrests,
one has to question the benefits of increasing the length of
time served.

RECOMMENDATION 2: Eliminate the use of prison for
parole or probation technical violators.
Today anywhere from 50-65% of the 650,000 prisoners
admitted to state prison are those who have failed to complete their terms of probation or parole. Of those who have
failed probation or parole, about half are being sent to prison
for what is known as technical violations. Such violations include such behavior as absconding from supervision, failure
to pay supervision fees, restitution costs, or fines, failure to
attend treatment, drug use as detected by urine-analysis,
failure to maintain employment, or being arrested (but not
convicted) for misdemeanor or felony level crimes.

While these behaviors are troublesome, they do not
reach the level of seriousness of requiring incarceration for
many months or years in state or federal prison. In many
ways using the most severe form of punishment for behavior
that only a probationer or parolee can be imprisoned for is
a clear example of where the punishment does not fit the
crime. For those that suggest that by incarcerating people
for non-felony crimes or non-compliance with the terms of
probation or parole we are preventing more serious crimes
to occur, there is no scientific data to support such a claim. In
fact as shown in recommendation No. 3, the evidence is that
parole supervision is either ineffective or criminogenic with
respect to public safety.
Parolees who are returned to prison because of technical violations often serve relatively long prison terms. The
U.S. Department of Justice reports that parole violators returned to prison serve an average of 18 months before being
re-released. In Louisiana, the state with the highest incarceration rate in the United States, the average length of stay
for a technical violator is 20 months. Conversely, the state
of Washington, by statute, does not allow technical violators
to spend more than 60 days incarcerated for such violations
and then only after a set number of violations.
These data clearly suggest that we are spending a great
deal of money and wasting a large amount of prison space
on people who fail to comply with parole and probation supervision rules and who have not committed new crimes.
Prosecutors and correctional officials erroneously believe
that unless individuals are re-incarcerated for technical violations, the individuals will commit serious crimes in the future.
There is no scientific evidence to support this belief and attendant policy, yet it continues to be the primary rationale
for re-incarcerating tens of thousands of people for criminal
or non-criminal behavior for which ordinary citizens could
not be incarcerated.

RECOMMENDATION 3: Reduce the length of parole
and probation supervision periods.

Table 10: Re-Incarceration Rates by Type of Release for Selected States
Release Type

Kentucky

Texas

Pennsylvania

Parole Supervision

53%

26%

50%

Discharges

18%

11%

19%

Total

35%

25%

42%

Source: James Austin, Patricia Handyman, and John Irwin. “Exploring the Needs and Risks of the Returning Prisoner Population.” (Presented to the From Prison to Home
Conference, Jan. 30-31, 2002).

23

Why and How to Reduce America’s Prison Population

Currently, persons placed on probation and parole remain in this status for extended periods of time (after three
years or more). Violation of the rigorous rules imposed on
probationers and parolees can result in return to prison. As
we saw in Table 4, a substantial fraction of prison admissions
are for these “technical” violations.
There is little evidence that lengthy parole and probation terms decrease crime. A number of studies in California
discovered that 1) there was no relationship between the
time on supervision and parole success, and 2) parole versus
no parole supervision on recidivism rates.70 Probation or parole supervision failure is most likely to occur within the first
12 months of supervision; thereafter, supervision is more of
a nuisance than a means for assisting people after prison or
preventing them from committing another crime.
Now, there is new research evidence indicating that parole supervision is largely ineffective with respect to reducing recidivism. Table 10 shows the recidivism rates by type of
release for Kentucky, Texas, and Pennsylvania. Here one can
see that people who complete their prison sentences without parole supervision have significantly lower recidivism
rates than those placed on parole supervision. The obvious
explanation is that people who have no parole obligations
when they leave prison (“max out”) can only be returned to
prison if they are convicted of a new felony crime. Conversely,
those placed on parole and probation can be re-incarcerated
for either non-criminal behavior or misdemeanor crimes. A
2005 Urban Institute study, among others, revealed that individuals released with no parole supervision return to prison
at a significantly lower rate than those released on parole.71

RECOMMENDATION 4: Decriminalize “victimless”
crimes, particularly those related to drug use and
abuse.
In recent years, behaviors have been criminalized that
are not dangerous and pose little if any threat to others. A
large group of people are currently serving time for behaviors that have been criminalized to protect people from
themselves. Their offenses involved the consent of all immediate parties to the transaction. Common examples in
American history have included abortion, gambling, illicit
sexual conduct that does not involve coercion (e.g., prostitution and, until recently, homosexual activity), and the sale

and possession of recreational drugs. According to the U.S.
Department of Justice, approximately 30-40% of all current
prison admissions involve crimes that have no direct or obvious victim other than the perpetrator (see Table 11).72 The
drug category constitutes the largest offense category, with
31% of all prison admissions resulting from such crimes.
In the late 19th and early 20th centuries, the United States
government conducted a large-scale experiment regarding
the social consequences of creating a new category of crimes
by making it illegal to distribute alcohol. Prohibition aimed
to stop people from drinking, and to an extent was successful. But the price of this success was ultimately considered
too high. Revenues from selling alcohol illegally swelled the
coffers of organized crime and magnified levels of corruption
in local governments. Gangsters gunned one another down
to gain control of the lucrative market for illegal liquor. Many
died from drinking alcohol put on the market without quality control. Eventually, Americans realized that Prohibition
was doing more harm than good and repealed it.
Politicians and the public have ignored the lessons
of Prohibition in formulating drug policy. In an attempt to
eliminate harms caused to individuals, their families and society from the abuse of heroin, cocaine, marijuana, and other
drugs, legislatures have passed laws sending large numbers
of users, abusers and low-level dealers to prison with very
long sentences. This prohibition, like the earlier one, has led
to violence as dealers have sought to eliminate rivals in the
lucrative market. Like Prohibition, it has resulted in the distribution of adulterated drugs that have injured and killed
users. The high profits of drug dealing are largely the consequence of legislation that eliminates competition from anyone unwilling to risk draconian penalties.
Every time a dealer is taken out of circulation by a prison
sentence, a new dealer is drawn in by the lure of large profits.
The prosecution and imprisonment of low-level traffickers
has increased racial disparities, and is the largest factor contributing to the rapid rise in imprisonment rates for women.
Dealers’ use of violence to eliminate competition helps to sustain the myth linking drug use to violence. Notwithstanding
our extraordinary effort to discourage the use and sale of illegal drugs, they remain widely available and widely used.
Though other Western nations have not decriminalized commerce in illegal drugs, they give greater weight to
medical and public health considerations in the formulation

Dorothy R. Jaman, Lawrence A. Bennett, and John E. Berecochea. Early Discharge from Parole: Policy, Practice and Outcome. Sacramento: California Department of
Sacramento Corrections, 1974; Deborah Star. Summary Parole: A Six and Twelve Month Follow-Up Evaluation. Sacramento: California Dept. of Corrections, Research Unit,
1979; Patrick G. Jackson. “Living Together Unmarried: Awareness Contexts and Social Interaction.” Journal of Family Issues (1983), and Patrick G. Jackson. “Bay Area Parole
Project.” Mimeographed, 1978.
71
Amy L. Solomon, Vera Kachnowski, and Avinash Bhati. Does Parole Work? Analyzing the Impact of Post-Prison Supervision on Re-Arrest Outcomes. Washington, DC: The
Urban Institute, 2005.
72
Criminological literature refers to these crimes as “victimless” (Edwin M. Schur. Crime Without Victims: Deviant Behavior and Public Policy. Englewood Cliffs, NJ: Prentice
Hall, 1965), suggesting that they do no harm, which is not always the case. For that reason we call these crimes “consensual.”
70

24

Unlocking America

Table 11: Most Serious Offense for Sentenced Prisoners, 2002 Prison Admissions
Most Serious Offense
Number of admissions

Percent

508,955

Violent offenses

27%

Property offenses

30%

Drug offenses

31%

Possession

10%

Trafficking

15%

Other/unspecified drug

7%

Public-order offenses

12%

Weapons

3%

Driving while intoxicated

3%

Other public-order

5%

Other offenses

1%

Source: Prison Statistics. US DOJ. 1 Aug. 2006 http://www.ojp.usdoj.gov/bjs/keyfacts.htm/

of drug policy. The violence that surrounds drug trafficking
in the United States is largely absent in those countries, and
governments are more receptive to needle exchange programs designed to limit the spread of AIDS.73
This does not mean that the government should simply walk away from the drug problem. It would be perfectly
appropriate for governments to conduct educational campaigns about drugs, for example. Regulatory approaches,
such as are now used for drugs that are not illegal should be
given serious consideration. The success of recent referenda
in several states allowing medical use of marijuana suggests
that the public opinion may be changing.74

TWO ADDITIONAL RECOMMENDATIONS THAT BEAR
ON HUMANE JUSTICE
In addition to the four recommendations to reduce the
prison population, there are two other recommendations
that need to be made that bear on the quality and fairness of
our use of imprisonment.

RECOMMENDATION 5: Improve conditions of
imprisonment.

Reducing the size of the prison population is not enough.
We could not in good conscience recommend that anyone
serve a prison sentence unless we ensure that those who do
end up behind walls are treated in a humane manner. Prisons
that systematically deny human dignity, basic human rights,
and life necessities are creating festering sores that poison
the entire society.
Unsafe, inhumane, and secretive prisons not only traumatize the incarcerated but also contaminate prison staff
and their families, as well as townsfolk near the prison. More
generally, support for an inhumane prison system requires
that prison workers and the public embrace the simplistic
concept that prisoners are unworthy beings that deserve
their harsh punishment above and beyond the segregation
from society and loss of freedom from incarceration itself.
The state can operate prisons efficiently and effectively while treating prisoners in a manner consistent with the
minimum standards and rights for prisoners formulated by
many private and public bodies in the 1960s and 1970s.75 In
the 1960s, the courts, particularly the U.S. Supreme Court, after virtually ignoring the plight of prisoners for decades, issued a number of decisions upholding prisoners’ rights and

73
Peter Reuter. “Why Can’t We Make Prohibition Work Better? Some Consequences of Ignoring the Unattractive.” Perspectives on Crime and Justice: 1996-1997 Lecture
Series, Vol. 1. Washington, DC: National Institute of Justice, 1997 http://www.ncjrs.gov/pdffiles/166609.pdf/
74
Between 1995 and the present, a majority of voters in state initiatives held in Alaska, Arizona, Colorado, Maine, Nevada, Oregon, the state of Washington, and Washington
D.C., voted to allow marijuana use in connection with medical treatment. In state and national polls, substantial majorities support legalizing marijuana for medical use. Details
are available at www.medicalmarijuanaprocon.org.
75
In 1955, the United Nations Congress on the Prevention of Crime and the Treatment of Prisoners adopted a set of Standard Minimum Rules for the Treatment of Prisoners.
See: Human Rights: A Compilation of International Instruments. New York: United Nations, 2002. Section G. In The Struggle for Justice: A Report on Crime and Punishment in
America. American Friends Service Committee. New York: Hill and Wang, 1971. 168-69, the Working Party for the American Friends Service Committee, which consisted of
individuals with a variety of prison-system experiences, produced one of the best-thought-out lists of humane conditions.

25

Why and How to Reduce America’s Prison Population

due process, and banning cruel and unusual punishment.
These decisions produced an array of mandated changes in
prisons across the country. However, after the mid-1970s,
the Supreme Court effectively returned to the “hands-off”
policy.76
There are five fundamental features of prison administration that should be acceptable to anyone interested in accomplishing the prison’s practicable purposes—punishment
and deterrence—without engaging in unnecessary, counterproductive, and cruel practices.
1. Cruel and Unusual Punishment
Prison overcrowding, the adoption of excessively
harsh and arbitrary control practices in reaction to prison
violence, and the growing general punitive attitude toward prisoners have resulted in more punitive policies
and practices. These include denial of adequate medical services, excessive use of physical force, and housing
prisoners in exceptionally punitive arrangements, such as
solitary confinement units and cells. The federal courts
have ruled that all of these practices violate the Eighth
Amendment of the United States Constitution.
2. Safety
Prisoners should be protected from assault, rape,
murder, etc., by other prisoners and staff. Effective strategies such as adequate surveillance, voluntary access to
safe living areas within the prison, housing prisoners in
small units, and “single-celling” should be practiced to ensure prisoners’ safety.
3. Health
Prisoners should have access to the resources and
services required to maintain their physical and mental
health. These include access to medical and psychiatric
services, adequate diet, and recreation. Prisoners should
not be subjected to physically and mentally deleterious
incarceration regimens such as extended periods of isolation and, restricted mobility, and excessive noise.
4. Programs
Any rational and humane system of punishment
should provide access to program opportunities that increase prison safety and improve prisoners’ chances of
making it in the community after prison. Such programs
would include academic, technical and citizenship education, as well as a wide variety of treatment programs
that help prisoners improve themselves and develop
more conventional, law-abiding interests and pursuits.
These types of programs should be provided regardless

76
77

of whether they reduce recidivism.
5. Post-Release Assistance
Most prisoners receive little or no preparation for
release from prison or assistance subsequent to their release. They face extraordinary difficulties in achieving stability, viability, and life fulfillment on the outside. States
should develop and provide access to transitional and
permanent housing, education, vocational training and
placement, counseling, coaching, and mentoring.

RECOMMENDATION 6: Restore ex-prisoner voting
and other rights.
Prisoners face exceptional problems from their stigmatized and reduced social and civil status. They are automatically barred from most city, county, and state employment
and from some housing such as federally subsidized housing
and are systematically denied employment by many private
employers. Their right to vote varies from state to state, even
from county to county in some states.
It will be difficult to overcome private employers’ restrictions on hiring ex-prisoners, but persistent efforts should be
made toward this goal. Perhaps laws against discrimination
against employment of ex-prisoners can be adopted in the
future. Government agencies, however, could easily change
their policies. In San Francisco, the city government has removed the question regarding prior arrests on job applications. Other government jurisdictions should follow this example. Opening up these relatively good-paying jobs to exprisoners greatly increases public safety by moving potential
criminals into conventional pathways. Government subsidies
for hiring ex-convicts could overcome some employers’ hesitations. In addition, exclusion from welfare, public housing,
and subsidies should be ended as should rules barring exconvicts from living in certain neighborhoods. Licensing restrictions should be maintained only when they are demonstrably necessary to protect the public.

ESTIMATED IMPACT OF THESE RECOMMENDATIONS
ON PRISON POPULATIONS
To illustrate what our recommendations can accomplish in reducing prison populations without impacting
crime rates and at considerable savings to taxpayers, we have
developed rough projections based on three of the four major recommendations for changing current sentencing and
correctional practices.:77

See: Jack E. Call. “The Supreme Court and Prisoners’ Rights.” Federal Probation (1995): 36-46, for a discussion of the Court’s shift in prisoners’ rights matters.
We are unable to estimate the effect of reducing the length of parole and probation supervision.

26

Unlocking America

Figure 5: Historical & Projected US Prision Population

1. Time served in prison would be reduced.
2. Technical parole and probation violators would not
serve time in prison for such behavior.
3. People convicted of “victimless” crimes would not be
sentenced to state prison.
We discuss below the implementation of these three
policy reforms briefly.

1. Time served in prison would be reduced.
Of the three recommendations, this one is clearly the
most important and most powerful in terms of reducing the
prison population. It also is the most acceptable to most politicians and the public as it does not eliminate incarceration
but make the amount of time served proportional to the se-

Table 12: Current and Projected State Prison Systems Based on Recommendations

Prison Admissions
Total
New Court Admissions
Probation Technical Violators
Parole Violators
Parole Technical Violators
Incarceration Rate

27

Current Practices
Time Served
Admits
Prisoners
(mos.)
650,000
27
1,446,250

422,500

New Practices
Time Served
(mos.)
19

Admits

Prisoners
666,250

390,000

30

975,000

292,500

21

511,875

65,000

30

162,500

32,500

21

56,875

195,000

19

308,750

97,500

12

97,500

97,500

19

154,375

0

0

0

483 per 100,000

222 per 100,0000

Why and How to Reduce America’s Prison Population

The length of imprisonment can be modestly reduced
(3-5 months) by immediately increasing the amount of good
time awarded to prisoners for good conduct and program
completion. Within states with indeterminate sentencing,
parole grant rates can be immediately increased especially
for prisoners who pose little risk to public safety. But ultimately, legislation will be needed to remove many of the restrictions that have served to increase the period of confinement (e.g., mandatory minimums, truth in sentencing, etc.).
Such reforms should be retroactive to the current prison
population.
2. Technical parole and probation violators would
not serve time in prison for such behavior.
The imprisonment of technical parole violators is a
clear example where the punishment does not fit the crime.
Further, we also know that recidivism rates are lowest for
persons who discharge from prison rather than facing a
lengthy period of parole supervision.
This reform can be implemented administratively by
not allowing revocations for such behavior. A number of
states have administratively implemented such reforms
including Michigan, Oregon, South Dakota, and Texas. As
noted earlier, the state of Washington passed legislation
more than 20 years ago that prohibits state imprisonment
for technical parole violators with no adverse impact on the
state’s crime rate. For those who are readmitted for a parole
violation, the period of re-confinement would be far shorter
than what it has been (no more than 90 days). This is consistent with recent legislation in Louisiana, which limits the
period of confinement for first time violators to 90 days and
in Washington where violators can serve no more than 60
days in a local jail. Similar to parole violators, few persons
should be sent to state or federal prison for anything that
does not constitute a conviction for a felony crime or for
non-criminal behavior.
3. People convicted of “victimless” crimes would
not be sentenced to prison.
Large numbers of persons arrested and convicted of
drug possession, disorderly conduct, public intoxication,
drunk driving, prostitution, curfew violation, vagrancy, loitering, gambling, and a wide variety of motor vehicle violations are being incarcerated in our jails and are being placed
on years of probation. Once placed on probation, they are
vulnerable to being sent to prison as probation violators
for continuing such behavior, failing to pay their probation
78

supervision fees, maintaining employment, attending treatment, or many other non-criminal acts.
Collectively, these three reforms, if implemented, would drop the prison population by over
50% and produce an incarceration rate of 222 per
100,000 people, which is what it was in 1986 and
which is still well above the rates that existed for
some 50 years before that. The prison population
would decline from 1.5 million to below 700,000 (see
Table 12, and Figure 5). Persons convicted of serious and violent crimes would continue to be incarcerated. But large numbers of probation and parole
violators and others now convicted of victimless
crimes would be diverted from state prison. All of
these reforms require no program funding—indeed
the adoption of such reforms would more than pay
for whatever prison and post-release programs that
would be of benefit to released prisoners. All of this
can be done without negatively impacting the crime
rate.
To those who say that these three basic recommendations are neither feasible nor practical we would simply note that these practices and laws are now in place
in many states and other countries. There are nine
states with incarceration rates that are near or well below the 222 per 100,000 population rate. The state of
Washington, by statute, does not allow parole violators
to be sent to prison. Louisiana recently passed legislation that greatly restricts first time technical violators
to be re-admitted to prison. Nevada recently passed
legislation that reduced the period of parole supervision, which has increased the parole success rate and
reduced the size and costs of the parole population.
In the area of decriminalization, 12 states (Alaska,
California, Colorado, Maine, Minnesota, Mississippi,
Nebraska, Nevada, New York, North Carolina, Ohio, and
Oregon) and several European countries along with
New Zealand and Australia have largely decriminalized
or reduced the penalties for drug possession.78 A number of local cities have also modified their local ordinances and criminal justice practices to decriminalize
pot (Berkeley, Oakland, and San Francisco, California;
Breckenridge, Colorado; Amherst, Massachusetts;
Madison and Milwaukee, Wisconsin; Urbana and
Carbondale, Illinois; and Colombia, Missouri). In all of
these jurisdictions there has been no associated increase in crimes rates.

Robert J. MacCoun and Peter Reuter. Drug War Heresies: Learning from Other Vices, Times, and Places. New York: Cambridge UP, 2001.

28

VI
Concluding Remarks
Reducing the number and length
of prison terms will require changes in
sentencing laws and parole and probation practices. This will not occur
until the public passes referenda and
successfully pressures legislators and executives, or enlightened political leaders
better understand the realities and move
on their own to make necessary changes.
There are a variety of methods and strategies to achieve this goal. No particular
political structure can guarantee or prevent the progress that is now needed.
Sentencing commissions, for example,
have been excellent devices for controlling over-incarceration in some states
(Oregon, Minnesota), while in other jurisdictions, such as the federal, sentencing
systems have been specifically designed
to incapacitate as many people as possible, rather than focusing on what offenders deserve.
We also recognize that as the system
of imprisonment has grown, so too has
the investment and the vested interests
that support its operations and growth.

29

In order to reverse the current trends we
will have to find a way to re-allocate the
money, political influence, and jobs that
the current system provides. This will
not be easy and it will take many years to
wean us off the excessive use of imprisonment.
Our first goal was to document the
negative and ineffective consequences of
mass incarceration in human, economic,
and public safety terms. Our second was
to offer one basic and simple recommendation that by itself would have a significant reduction in the prison population
-- shorter periods of imprisonment that
are proportional to the harm inflicted
upon society and individuals. We hope
this report will stimulate a serious debate
on the use of imprisonment and lead to a
new policy of decarceration. If this would
occur, we could re-invest some portion of the tens of billions of dollars we
spend each year incarcerating millions of
Americans into those communities and
families that are now being unfairly devastated by imprisonment.

30

About the Authors
JAMES AUSTIN is the President of the JFA Institute. Prior to that, he was the Director
of the Institute of Crime, Justice and Corrections at the George Washington University, and
Executive Vice President for the National Council on Crime and Delinquency. Dr. Austin received
his Ph.D. in sociology from the University of California, Davis. He began his career in corrections
with the Illinois Department of Corrections in 1970 at Statesville Penitentiary. He was named by
the American Correctional Association as its recipient of the Peter P. Lejins Research Award, and
received the Western Society of Criminology Paul Tappin award for outstanding contributions in
the field of criminology. He also has served as the Chair of the National Policy Council for the
American Society of Criminology. In 2007 he was appointed to the California Department of
Corrections and Rehabilitation Expert Panel on Adult Offender Recidivism Reduction Programs.
He is the co-author of Its About Time: America’s Imprisonment Binge (with John Irwin).
TODD CLEAR is Distinguished Professor, John Jay College of Criminal Justice, City
University of New York. He received a Ph.D. in Criminal Justice from The University at Albany.
Clear has also held professorships at Ball State University, Rutgers University, and Florida State
University (where he was also Associate Dean of the School of Criminology and Criminal Justice).
This year he was elected president of The American Society of Criminology. His work has been
recognized through several awards, including those of the American Society of Criminology, the
Academy of Criminal Justice Sciences, The Rockefeller School of Public Policy, the American
Probation and Parole Association, the American Correctional Association, and the International
Community Corrections Association. Dr. Clear is author of Imprisoning Communities: How
Mass Incarceration Makes Disadvantaged Communities Worse (Oxford University
Press, 2007)
TROY DUSTER is Silver Professor of Sociology and Director of the Institute for the History
of the Production of Knowledge at New York University. Professor Duster received his Ph.D. in
sociology from Northwestern University. He also holds an appointment as Chancellor’s Professor
at the University of California, Berkeley. He is the past-president of the American Sociological
Association and a member of the Board of Advisors of the Social Science Research Council. He
is the former Director of the Institute for the Study of Social Change, both at the University of
California, Berkeley.
DAVID F. GREENBERG is Professor of Sociology at New York University. He received
his Ph.D. in physics from the University of Chicago. He is the author, co-author or editor of
The University of Chicago Graduate Problems in Physics, with Solutions,
Mathematical Criminology, Crime and Capitalism, Linear Panel Analysis: Models
of Quantitative Change, The Construction of Homosexuality, and “Criminal”
Careers, as well as dozens of journal articles. His research focuses on crime and criminal
justice, law, deviance and social control, human sexuality, statistical methods and mathematical
modeling, computational linguistics and Assyriology. He (along with John Irwin) was a contributing
author to the American Friends Service Committee’s influential report Struggle for Justice: A

31

Report on Crime and Punishment in America. Professor Greenberg is a Fellow of the American Society of Criminology.
JOHN IRWIN is Professor Emeritus at San Francisco State University where he taught sociology for nearly three decades.
After serving a five sentence for armed robbery in California’s prison system, he received his Ph.D. in sociology from the University of
California, Berkeley. His considerable research on prisons and sentencing has been published in five books: The Felon, Prisons
in Turmoil, The Jail, It’s About Time: America’s Imprisonment Binge (with James Austin), and The Warehouse
Prison. He (along with David Greenberg) was a contributing author to the American Friends Service Committee’s influential report
Struggle for Justice: A Report on Crime and Punishment in America. As an organizer and leader of the Prisoners’ Union
in California, he worked closely with the California legislature on the Uniform Sentencing Act passed in 1976. He received the August
Volmer award from the American Society of Criminology for outstanding contributions to criminal justice.
CANDACE MCCOY holds an appointment at the Graduate Center of the City University of New York and teaches in the
doctoral program of the John Jay College of Criminal Justice. Dr. McCoy received a J.D. from the University of Cincinnati and a
Ph.D. in Jurisprudence and Social Policy from the University of California and is a member of the Ohio bar. She specializes in the
study of criminal justice policies, researching and teaching on such topics as sentencing, plea bargaining, jury decision-making, and
police practices. Her most recent publication on these matters is “Plea Bargaining as Coercion: The Trial Penalty and Plea Bargaining
Reform,” The Criminal Law Quarterly, Vol. 50, No. 1 (2005). Professor McCoy has conducted many evaluations of innovative
programs and has consulted widely with federal and state criminal justice agencies most recently as Chair of New Jersey’s Criminal
Disposition Commission, and is currently working with the Scottish Centre for Crime and Justice Research to develop restorative justice
practices as conditions of probation. She has received the American Society of Criminology’s Herbert Block Award for distinguished
service to the profession.
ALAN MOBLEY is an assistant professor in the School of Public Affairs at San Diego State University. He received his Ph.D.
in Criminology, Law & Society from the University of California, Irvine. While a prisoner in the Federal Bureau of Prisons he earned
Master’s (Sociology, Vermont College, 1994) and Bachelor’s (Economics, SUNY, 1991) degrees. Dr. Mobley is a founding member of
All Of Us Or None, an organizing initiative to guarantee civil rights for former prisoners. He sits on the editorial board of The Journal of
Prisoners on Prisons, and is a board member and Past-President of Visions for Prisons, a nonprofit agency offering support services
to prisoners and prison staff. His current action-research focuses on community and restorative justice, peacemaking and multistakeholder dialog, and community based service-learning. He directs several grant funded projects in these areas.
BARBARA OWEN is a nationally-known expert in the areas of girls, women and crime, women-centered policy and women’s
prison culture. A Professor of Criminology at California State University, Fresno, she received her Ph.D. in sociology from the
University of California, Berkeley in 1984. Barbara Owen has written extensively on issues confronting women enmeshed in the
criminal justice system. She is the author of In the Mix, an ethnography on women’s prison culture and co-author of Genderresponsive Strategies: Research, Practice and Guiding Principles for Women Offenders, which received the
University of Cincinnati award for contributions to correctional practice. She regularly consults for state and federal correctional
agencies on such gender related issues. Her work was most recently honored with the Saltzman Award by the Division of Women and
Crime, American Society of Criminology.
JOSHUA PAGE is an assistant professor of sociology at the University of Minnesota. He received his Ph.D. from UC Berkeley,
where he published research on prison education and conducted an in-depth study of the California Correctional Peace Officers
Association (CCPOA). Professor Page is currently writing a book on the CCPOA and the politics of punishment in California. Further,
he is directing a study on the transition of youthful offenders from juvenile correctional facilities into the community, identifying factors
that facilitate or obstruct the youths’ ability to live successful, crime-free lives. He is also gathering survey data on the attitudes,
demographic composition, and working conditions of prison officers in Minnesota.
32